sox_ng/soxeffect_ng.7

'\" t
'\" The line above instructs most `man' programs to invoke tbl
'\"
'\" Separate paragraphs; not the same as PP which resets indent level.
.de SP
.if t .sp .5
.if n .sp
..
'\"
'\" Replacement em dash for nroff (default is too short).
.ie n .ds m " - 
.el .ds m \(em
'\"
'\" Placeholder macro for if longer nroff arrow is needed.
.ds RA \(->
'\"
'\" Decimal point set slightly raised
.if t .ds d \v'-.15m'.\v'+.15m'
.if n .ds d .
'\"
'\" Enclosure macro for examples
'\" From groff-1.23 CW is undeclared by default but nroff is CW anyway
.if n .ftr CW R
.de XE
.SP
.nf
.ft CW
..
.de XX
.ft R
.SP
.fi
..
.TH SOXEFFECT_NG 7 "December 05, 2024" "SoX" "Sound eXchange_ng"
.SH NAME
soxeffect_ng \- Effects supported by sox_ng and libsox_ng
.SH SYNOPSIS
In addition to converting, playing and recording audio files, SoX can
be used to invoke a number of audio effects.  Multiple effects may
be applied by specifying them one after the other at the end of the SoX
command line, forming an `effects chain'.
Note that applying multiple effects in real time (i.e. when playing audio)
may require a high performance computer.
.SP
Some of the SoX effects are primarily intended to be applied to a single
instrument or `voice'. To facilitate this, the \fBremix\fR effect and
the global SoX option \fB\-M\fR can be used to isolate then recombine
tracks from a multitrack recording.
.SH MULTIPLE EFFECTS CHAINS
A single effects chain is made up of one or more effects.  Audio from
the input runs through the chain until either the end of the input file
is reached or an effect terminates the chain.
.SP
SoX supports running multiple effects chains over the input audio.
In this case, when one chain indicates that it is done processing audio,
the audio data is sent through the next effects chain.  This
continues until either no more effects chains exist or the input has
reached the end of the file.
.SP
Effects chains can be separated by placing a
.B :
(colon) after an effect;
any following effects are a part of a new effects chain.
.SP
It is important to place the effect that stops the chain
as the first effect in the chain because any samples
that are buffered by effects to the left of the terminating effect
will be discarded.  The amount of samples discarded is related to the
.B \-\-buffer
option and it should be kept small, relative to the sample rate, if
the terminating effect cannot be first.  Further information on
stopping effects can be found in the
.B Stopping SoX
section.
.SP
There are a few pseudo-effects that can help when using multiple effects chains.
These include
.BR newfile ,
which starts writing to a new output file before moving to the
next effects chain, and
.BR restart ,
which moves back to the first effects chain.  Pseudo-effects
must be specified as the first effect in a chain and as the only
effect in a chain (i.e. they must have a
.B :
before and after them).
.SP
Here is an example of multiple effects chains.  It splits the
input file into multiple files, each of 30 seconds in length and
each output filename will have unique number in its name,
as documented in the
.B Output Files
section.
.XE
   sox_ng in.au out.au trim 0 30 : newfile : restart
.XX
.SH COMMON NOTATION AND PARAMETERS
In the descriptions that follow,
[square brackets] are used to denote parameters that are optional,
{braces} to denote those that are both optional and repeatable,
<angle brackets> to denote those that are repeatable but not
optional and pipe characters `|' separate options from which to choose
one of several alternatives.
Where applicable, default values for optional parameters are shown
(in parentheses).
.SP
The following parameters are used with, and have the same meaning for,
several effects:
.TP
\fIfrequency\fR
A frequency in Hz or, if followed by \fBk\fR, in kHz or,
if preceded by \fB%\fR, in semitones relative to A (440Hz);
alternatively, scientific note names (e.g. E2) may be used.
.TP
\fIgain\fR
A power gain in dB.
Zero gives no gain, less than zero gives an attenuation
and greater than zero amplifies.
.TP
\fIduration\fR
See \fBTime Specifications\fR below.
.TP
\fIposition\fR
A position within the audio stream; the syntax is
[\fB=\fR\^|\^\fB\-\fR\^|\^\fB+\fR]\fItimespec\fR, where \fItimespec\fR is a
time specification (see below). The optional first character indicates
whether the \fItimespec\fR is to be interpreted relative to the start
(\fB=\fR) or end (\fB\-\fR) of the audio or relative to the previous
\fIposition\fR (\fB+\fR) if the effect accepts multiple positional arguments.
The audio length
must be known for end-relative locations to work, though some effects do accept
\fB\-0\fR for the end of the audio even if the length is unknown.
Which of \fB=\fR, \fB\-\fR and \fB+\fR is the default depends on the effect
and is shown in its syntax as, e.g., \fIposition\fR(+).
.SP
Examples: `=2:00' is two minutes into the audio stream,
`\-100s' is one hundred samples before the end of the audio,
`+0:12+10s' is twelve seconds and ten samples after the previous position
and
`\-0.5+1s' is one sample less than half a second before the end of the audio.
.TP
\fIwidth\fR[\fBh\fR\^|\^\fBk\fR\^|\^\fBo\fR\^|\^\fBq\fR]
Used to specify the bandwidth of a filter.  A number of different
methods to specify the width are available (though not all for every effect).
One of the characters shown may be appended to select the desired method
as follows:
.SP
.ne 5
.TS
center;
cI cI lI
cB c l.
\ 	Method	Notes
h	Hz	\ 
k	kHz	\ 
b	Hz	Old non-frequency-warped response
o	octaves	\ 
q	Q-factor	See [2]
s	slope	\ 
.TE
.SP
For each effect that uses this parameter, the default method (if no
character is appended) is the one that is listed first in the first line of
the effect's description.
.SH TIME SPECIFICATIONS
A \fItimespec\fR can be given in one the following two forms:
.TP
[[\fIhours\fB:\fR]\fIminutes\fB:\fR]\fIseconds\fR[\fB.\fIfrac\fR][\fBt\fR]
For example, a time specification of `1:30.5' corresponds to
one minute, thirty and \(12 seconds.
The component values do not have to be normalized; e.g.
`1:23:45', `83:45', `79:0285', `1:0:1425', `1::1425' and `5025'
are all equivalent.
.TP
\fIsamples\fBs\fR
Specifies the number of samples directly, as in `8000s'.  For large sample
counts, \fIe notation\fR is supported: `1.7e6s' is the same as `1700000s'.
.PP
Time specifications can also be chained with \fB+\fR or \fB\-\fR into a new
time specification where the right part is added to or subtracted from the
total so far.
For example, `3:00\-200s' means two hundred samples less than three minutes.
.PP
If a \fItime specification\fR is a plain whole number with no \fBt\fP or \fBs\fR
suffix, whether it is taken as a number of seconds or a number of samples
depends on the effect in question. At present, it always means seconds
except for the \fIduration\fR parameters of the \fBsilence\fP effect.
.SH SUPPORTED EFFECTS
To see whether SoX has support for an optional effect, enter
.B sox_ng \-h
and look for its name in the \fBEFFECTS\fR list;
a categorized list of the effects can be found in the
accompanying README file.
.TP
\fBallpass\fR [\fB\-1\fR\^|\^\fB2\fR] \fIfrequency \fR[\fIwidth\fR[\fBh\fR\^|\^\fBk\fR\^|\^\fBo\fR\^|\^\fBq\fR]]
Apply a two-pole all-pass filter with central frequency \fIfrequency\fR
and filter width \fIwidth\fR.
An all-pass filter changes the
audio's frequency to phase relationship without changing its frequency
to amplitude relationship.  The filter is described in detail in [1].
.SP
\fB-1\fR or \fB-2\fR use an experimental 1-pole or 2-pole filter,
in which case \fIwidth\fR does not apply.
.SP
This effect supports the \fB\-\-plot\fR global option.
.TP
\fBband\fR [\fB\-n\fR] \fIfrequency\fR [\fIwidth\fR[\fBh\fR\^|\^\fBk\fR\^|\^\fBo\fR\^|\^\fBq\fR]]
Apply a band-pass filter.
The frequency response drops logarithmically
around the center
.IR frequency .
The
.I width
parameter gives the slope of the drop:
the frequencies at
.I frequency
+
.I width
and
.I frequency
\-
.I width
will have half their original amplitudes.
Its default value is half of the center frequency.
.SP
.B band
defaults to a mode oriented to pitched audio,
i.e. voice, singing or instrumental music.
The \fB\-n\fR (for noise) option uses the alternate mode
for unpitched audio (e.g. percussion),
though
\fB\-n\fR introduces a power gain of about 11dB in the filter, so beware
of output clipping.
.B band
introduces noise in the shape of the filter,
peaking at the
center
frequency and settling around it.
.SP
This effect supports the \fB\-\-plot\fR global option.
.SP
See \fBsinc\fR for a band-pass filter with steeper shoulders.
.TP
\fBbandpass\fR\^|\^\fBbandreject\fR [\fB\-c\fR] \fIfrequency width\fR[\fBh\fR\^|\^\fBk\fR\^|\^\fBo\fR\^|\^\fBq\fR\^|\^\fBb\fR]
Apply a two-pole Butterworth band-pass or band-reject filter with
central frequency \fIfrequency\fR, and (3dB-point) bandwidth \fIwidth\fR.
The
.B \-c
option applies only to
.B bandpass
and selects a constant skirt gain (peak gain = Q) instead of the
default, a constant 0dB peak gain.
The filters roll off at 6dB per octave (20dB per decade)
and are described in detail in [1].
.SP
These effects support the \fB\-\-plot\fR global option.
.SP
See \fBsinc\fR for a band-pass filter with steeper shoulders.
.TP
\fBbass\fR\^|\^\fBtreble \fIgain\fR [\fIfrequency\fR [\fIwidth\fR[\fBs\fR\^|\^\fBh\fR\^|\^\fBk\fR\^|\^\fBo\fR\^|\^\fBq\fR]]]
Boost or cut the bass (lower) or treble (upper) frequencies of the audio
using a two-pole shelving filter with a response similar to that
of a standard hifi's tone controls.  This is also
known as shelving equalization.
.SP
\fIgain\fR gives the gain at 0Hz for \fBbass\fR or, for \fBtreble\fR,
whichever is the lower of \(ap22kHz and the Nyquist frequency.
Its useful range is about \-20 (for a large cut) to +20 (for a large boost).
Beware of
.B Clipping
when using a positive \fIgain\fR.
.SP
The filter can be fine-tuned using the following optional parameters:
.SP
\fIfrequency\fR sets the filter's central frequency and so can be
used to extend or reduce the frequency range to be boosted or
cut.
The default values are 100Hz for \fBbass\fR and 3kHz for \fBtreble\fR.
.SP
\fIwidth\fR
determines how
steep the filter's shelf transition is.  In addition to the common
width specification methods,
`slope' (the default) may be used.
Its useful range is
about 0\*d3 for a gentle slope to 1 (the maximum) for a steep slope;
and its default value is 0\*d5.
.SP
The filters are described in detail in [1].
.SP
These effects support the \fB\-\-plot\fR global option.
.SP
See \fBequalizer\fR for a peaking equalization effect.
.TP
\fBbend\fR [\fB\-f \fIframe-rate\fR(25)] [\fB\-o \fIoversampling\fR(16)]
{\fIstart-position(+)\fB,\fIcents\fB,\fIend-position(+)\fR}
.SP
Changes the pitch by specified amounts at specified times without changing
the duration.
Each given triple: \fIstart-position\fB,\fIcents\fB,\fIend-position\fR
specifies one bend.
\fIcents\fR is the number of cents (100 cents = 1 semitone) by which to
bend the pitch. The other values specify the points in time at which to start
and end bending the pitch.
During each bend, the frequency changes logarithmically,
i.e. by the same number of cents per second.
.SP
The pitch bending algorithm uses the Discrete Fourier Transform (DFT)
at a particular frame rate and oversampling rate.
The
.B \-f
(from 10 to 80)
and
.B \-o
(from 4 to 32)
parameters may be used to adjust these parameters and thus control the
smoothness of the changes in pitch.
.SP
For example, an initial tone is generated, then bent three times, yielding
four different notes in total:
.XE
.ne 2
   play_ng \-n synth 2.5 sin 667 gain 1 \e
	bend .35,180,.25  .15,740,.53  0,\-520,.3
.XX
Here, the first \fBbend\fR runs from 0.35 to 0.6 seconds and the second one from 0.75
to 1.28 seconds.
Note that the clipping that is produced in this example is deliberate;
to remove it, use
.B gain\ \-5
in place of
.BR gain\ 1 .
.SP
See \fBpitch\fR.
.TP
\fBbiquad \fIb0 b1 b2 a0 a1 a2\fR
Apply a biquad Infinite Impulse Response filter
with the given coefficients, where \fIb*\fR and \fIa*\fR are
the numerator and denominator coefficients respectively.
.SP
See http://en.wikipedia.org/wiki/Digital_biquad_filter (where a0 = 1).
.SP
This effect supports the \fB\-\-plot\fR global option.
.TP
\fBcentercut \fR[\fB\-a\ \fIgain\fR] [\fB\-b\fR] [\fB\-w \fIsize\fR]
Remove the center from a stereo file leaving the far left and right parts of
the stereo file intact and the center in a third channel.
.SP
The \fB\-a\fR option is \fIgain-out\fR for all channels, default 1\*d0..
.SP
The \fB\-b\fR option moves the bass (below 200Hz) out of the center
into the sides for those who want karaoke.
.SP
The \fB\-w\fR option changes the window size from its default of 8192
sample frames to a power of two from 4 to 32768.
.SP
A stereo equivalent of the \fBoops\fR effect is
.XE
    sox_ng in.wav centercut remix 1 2
.XX
.SP
The keymap for the \fB\-a\fR option is \fBcentercut.gain\fR.
.SP
See also \fBoops\fR.
.TP
\fBchannels \fIchannels\fR
Invoke a simple algorithm to change the number of channels in
the audio signal to the given number:
mixing if decreasing the number of channels or duplicating if
increasing the number of channels.
.SP
The
.B channels
effect is invoked automatically if SoX's \fB\-c\fR option specifies a
number of channels that is different to that of the input file(s).
Alternatively, if this effect is given explicitly, SoX's
.B \-c
option need not be given.  For example, the following two commands are
equivalent:
.XE
.ne 2
   sox_ng input.wav \-c 1 output.wav bass \-b 24
   sox_ng input.wav      output.wav bass \-b 24 channels 1
.XX
though the second form is more flexible as it allows the effects to
be ordered arbitrarily.
.SP
For example, when making a stereo file quadraphonic,
the left and right channels are copied into the third and fourth and
when mixing a four-channel file down to stereo, the left channel is
the mix of the first and third and the right of the second and fourth.
.SP
See
.B remix
for an effect that allows channels to be mixed and selected arbitrarily.
.TP
\fBchorus [\fB\-n\fR\^|\^\fBl\fR\^|\^\fBq\fR] [\fB\-s\fR\^|\^\fBt\fR] [\fIgain-in [gain-out\fR {\fIdelay [decay [speed [depth [\fB\-s\fR\^|\^\fBt\fR]]]]}]]
Add a chorus effect to the audio.  This can make a single voice sound
like a chorus but can also be applied to instrumentation.
.SP
Chorus resembles an \fBecho\fR effect with a short delay but,
while \fBecho\fR's delay is constant, \fBchorus\fR' delay
varies by a sinusoidal or triangular modulation.
.SP
See [3] for further discussion of the chorus effect.
.SP
The \fB\-l\fR flag makes \fBchorus\fR do linear interpolation between samples
when the offset into the delay line is not a whole number,
which is about 15% slower but makes it considerably less noisy and
\fB\-q\fR asks for quadratic interpolation which is
about 40% slower but makes it even less noisy.
\fB\-n\fR explicitly asks for no interpolation, the default, fast and fuzzy.
.SP
\fB\-s\fR or \fB\-t\fR before the stages change the default wave type
for all of them.
.SP
All parameters are all optional and, if missing, assume the following values:
.SP
.TS
lb lb lb lb
l c c le.
Parameter	Range	Default	Description
gain-in	-1-1	0.5	T{
Proportion of input
.br
delivered clean to the adder
T}
gain-out	-1-1	1	Final volume adjustment
delay	0-1000	40-60	Fixed delay in milliseconds
decay	-1-1	0.5	Volume of delayed output
speed	0-192k	0.25	Modulation frequency
depth	0-1000	2	Extra delay in milliseconds
wave	-s\^|\^-t	-s	Sinusoidal/triangular modulation
.TE
.SP
There are keymaps for \fBgain_in\fR and \fBgain_out\fR.
.SP
Each delay ranges from the fixed \fIdelay\fR to \fIdelay\fR\ \(pl\ \fIdepth\fR.
.SP
\fIGain-out\fR is then applied to the sum of the input scaled by \fIgain-in\fP
and the outputs from the delays scaled by their \fIdecay\fRs.
.SP
For internal reasons regarding the speed of \fBchorus\fR,
there is a limit of 256 chorus stages.
.SP
A typical delay is around 40ms to 60ms; the modulation speed is best
near 0\*d25Hz and the modulation depth around 2ms.
For example, a single delay:
.XE
   play_ng guitar1.wav chorus 0.7 0.9 55 0.4 0.25 2 \-t
.XX
Two delays of the original samples:
.XE
.ne 2
   play_ng guitar1.wav chorus 0.6 0.9 50 0.4 0.25 2 \-t \e
	 60 0.32 0.4 1.3 \-s
.XX
A fuller-sounding chorus (with three additional delays):
.XE
.ne 2
   play_ng guitar1.wav chorus 0.5 0.9 50 0.4 0.25 2 \-t \e
	 60 0.32 0.4 2.3 \-t 40 0.3 0.3 1.3 \-s
.XX
\fBflanger\fR can do everything that \fBchorus\fR does except multiple stages
but works in floating point internally instead of integers so is slower
without a floating point processor:
.XE
   chorus -l gain-in gain-out delay decay speed depth -wave
.XX
is equivalent to
.XE
   flanger delay depth 0 100\(mudecay\(digain-in speed wave 0 \e
      vol gain-out\(di(gain-in\(pldecay)
.XX
For a flow diagram of how \fBchorus\fR works,
say \fBsox_ng \-h chorus\fR.
.TP
\fBcompand \fIattack1\fB,\fIdecay1\fR{\fB,\fIattack\fB,\fIdecay\fR}
[\fIsoft-knee-dB\fB:\fR]\fIin-dB1\fR[\fB,\fIout-dB1\fR]{\fB,\fIin-dB\fB,\fIout-dB\fR}
.br
[\fIgain\fR [\fIinitial-volume-dB\fR [\fIdelay\fR]]]
.SP
Compand (compress or expand) the dynamic range of the audio.
.SP
The
.I attack
and
.I decay
parameters (in seconds) determine the time over which the
instantaneous level of the input signal is averaged to determine its
volume; attacks refer to increases in volume and decays refer to
decreases.
For most situations, the attack time (its response to the music getting
louder) should be shorter than the decay time because the human ear is more
sensitive to sudden loud music than sudden soft music.
When more than one pair of attack/decay parameters is specified,
each input channel is companded separately and the number of
pairs must agree with the number of input channels.
Typical values are 0\*d3,0\*d8 seconds.
.SP
The second parameter is a list of points on the compander's transfer
function specified in dB relative to the maximum possible signal
amplitude.  The input values must be in a strictly increasing order but
the transfer function does not have to be monotonically rising.  If
omitted, the value of
.I out-dB1
defaults to the same value as
.IR in-dB1 ;
levels below
.I in-dB1
are not companded but may have gain applied to them.
The point `0,0' is assumed but may be overridden by
`0,\fIout-dBn\fR'.
If the list is preceded by a
.I soft-knee-dB
value, then the points at where adjacent line segments on the
transfer function meet are rounded by the amount given.
Typical values for the transfer function are `6:\-70,\-60,\-20'.
.SP
The third (optional) parameter is an additional gain in dB to be applied
at all points on the transfer function and allows easy adjustment
of the overall gain.
.SP
The fourth (optional) parameter is an initial level to be assumed for
each channel when companding starts.  This lets you supply a
nominal level initially so that, for example, a very large gain is not
applied to initial signal levels before the companding action has begun
to operate: it is quite probable that in such an event, the output would
be severely clipped while the compander gain adjusts itself.
A typical value (for audio which is initially quiet) is
.B \-90
dB.
.SP
The fifth (optional) parameter is a delay in seconds.  The input signal
is analyzed immediately to control the compander, but it is delayed
before being fed to the volume adjuster.  Specifying a delay
approximately equal to the attack/decay times allows the compander to
operate in a predictive rather than a reactive mode.
A typical value is 0\*d2 seconds.
.SP
.TS
center;
c8 c8 c.
*	*	*
.TE
.SP
The following example might be used to make a piece of music with both
quiet and loud passages suitable for listening to in a noisy environment
such as a moving vehicle:
.XE
   sox_ng asz.wav asz-car.wav compand 0.3,1 6:\-70,\-60,\-20 \-5 \-90 0.2
.XX
The transfer function (`6:\-70,...') says that very soft sounds (below
\-70dB) remain unchanged.  This stops the compander from
boosting the volume on `silent' passages such as between movements.
However, sounds in the range \-60dB to 0dB (maximum
volume) are boosted so that the 60dB dynamic range of the
original music is compressed 3-to-1 into a 20dB range, which is
wide enough to enjoy the music but narrow enough to get around the
road noise.  The `6:' selects 6dB soft-knee companding.
The \-5\ dB output gain is needed to avoid clipping (the number is
inexact and was derived by experimentation).
The \-90\ dB for the initial volume will work fine for a clip that starts
with near silence and the delay of 0\*d2 seconds makes
the compander react more quickly to sudden volume changes.
.SP
In the next example, \fBcompand\fR is used as a noise-gate for when the
noise is at a lower level than the signal:
.XE
   play_ng in.au compand .1,.2 \-inf,\-50.1,\-inf,\-50,\-50 0 \-90 .1
.XX
Here is another noise-gate, this time for when the
noise is at a higher level than the signal (making it, in some ways,
similar to a squelch effect):
.XE
   play_ng in.au compand .1,.1 \-45.1,\-45,\-inf,0,\-inf 45 \-90 .1
.XX
This effect supports the \fB\-\-plot\fR global option (for the transfer function).
.SP
For a flow diagram of how \fBcompand\fR works,
say \fBsox_ng \-h compand\fR.
.SP
See
.B mcompand
for a multiple-band companding effect.
.TP
\fBcontrast \fR[\fIamount\fR(75)]
Comparable with compression, this effect modifies an audio signal to
make it sound louder.
.I amount
controls the amount of the enhancement and is a number in the range 0\-100.
Note that
.I amount
= 0 still gives a significant contrast enhancement.
.SP
There is a keymap for \fIamount\fR.
.SP
See the
.B compand
and
.B mcompand
effects.
.TP
\fBdcshift \fIshift\fR [\fIlimiter\-gain\fR]
Apply a DC shift to the audio.  This can be useful to remove a known DC
offset (caused perhaps by a hardware problem in the recording chain)
from the audio.  The effect of a DC offset is reduced headroom and
hence volume.
The
.B stat
or
.B stats
effect can be used to determine if a signal has a DC offset.
.SP
The given \fIdcshift\fR value is a floating point number in the range
of \(+-2 that indicates the amount to shift the audio (which is in the
range of \(+-1).
.SP
An optional
.I limiter\-gain
can be specified as well.  It should have a value much less than 1
(e.g. 0\*d05 or 0\*d02) and is used only on peaks to prevent clipping.
.SP
An alternative approach to removing a DC offset (albeit with a short delay)
is to use the
.B highpass
filter effect at a frequency of say 10Hz, as illustrated in the following
example:
.XE
   sox_ng \-n dc.wav synth 5 sin %0 50
   sox_ng dc.wav fixed.wav highpass 10
.XX
.TP
\fBdeemph\fR
Apply Compact Disc (IEC 60908) de-emphasis with a treble attenuation shelving
filter.
.SP
Pre-emphasis was applied in the mastering of some CDs issued in the early
1980s.  These included many classical music albums, as well as now
sought-after issues of albums by The Beatles, Pink Floyd and others.
Pre-emphasis should be removed at playback time by a de-emphasis
filter in the playback device.  However, not all modern CD players have
this filter and very few PC CD drives have it; playing pre-emphasized
audio without the correct de-emphasis filter results in audio that sounds harsh
and is far from what its creators intended.
.SP
With the
.B deemph
effect, it is possible to apply the necessary de-emphasis to audio that
has been extracted from a pre-emphasized CD and then either burn the
de-emphasized audio to a new CD (which will then play correctly on any
CD player) or simply play the correctly de-emphasized audio files on the
PC.  For example:
.XE
   sox_ng track1.wav track1\-deemph.wav deemph
.XX
and then burn track1-deemph.wav to CD, or
.XE
   play_ng track1\-deemph.wav
.XX
or simply
.XE
   play_ng track1.wav deemph
.XX
The de-emphasis filter is implemented as a biquad and requires the input
audio sample rate to be either 44.1kHz or 48kHz.  Its maximum deviation
from the ideal response is only 0\*d06dB (up to 20kHz).
.SP
This effect supports the \fB\-\-plot\fR global option.
.TP
\fBdelay\fR {\fIposition(=)\fR}
Delay zero or more audio channels such that they start at the given
\fIposition\fR.
.SP
For example,
.B delay 1\*d5 +1 3000s
delays the first channel by 1\*d5 seconds, the second channel by 2\*d5
seconds (one second more than the previous channel), the third channel
by 3000 samples and leaves other channels undelayed.
The following (one long) command plays a chime sound:
.XE
.ne 3
   play_ng \-n synth \-j 3 sin %3 sin %\-2 sin %\-5 sin %\-9 \e
	sin %\-14 sin %\-21 fade h .01 2 1.5 delay \e
	1.3 1 .76 .54 .27 remix \- fade h 0 2.7 2.5 norm \-1
.XX
and this an arpeggiated guitar chord:
.XE
.ne 2
   play_ng \-n synth pl G2 pl B2 pl D3 pl G3 pl D4 pl G4 \e
	delay 0 .05 .1 .15 .2 .25 remix \- fade 0 4 .1 norm \-1
.XX
With no parameters it does nothing.
To delay all channels by the same amount, use the \fBpad\fP effect.
.TP
\fBdither\fR [\fB\-S\fR\^|\^\fB\-s\fR\^|\^\fB\-f \fIfilter\fR] [\fB\-a\fR] [\fB\-p \fIprecision\fR]
Apply dithering to the audio.
Dithering deliberately adds a small amount of noise to the signal in
order to mask audible quantization effects that can occur if the output
sample size is less than 24 bits.  With no options, this effect
adds TPDF white noise.
.SP
The
.B \-S
option selects a slightly `sloped' TPDF, biased towards higher
frequencies.  It can be used at any sampling rate but, below \(~~22kHz,
plain TPDF is probably better and, above \(~~ 37kHz, noise-shaping
(if available) is probably better.
.SP
The \fB\-s\fR option enables noise-shaping with the \fBshibata\fR filter
(the same as \fB\-f\ shibata\fR)
and with the
.B \-f
option it is possible to select a particular noise-shaping filter from
the following list:
.nh
\fBlipshitz\fR, \fBf-weighted\fR, \fBmodified-e-weighted\fR,
\fBimproved-e-weighted\fR, \fBgesemann\fR, \fBshibata\fR, \fBlow-shibata\fR,
\fBhigh-shibata\fR, \fBshibata-\fR(\fBA\fR\^|\^\fBB\fR)(\fB0\fR\^|\^\fB1\fR\^|\^\fB2\fR\^|\^\fB3\fR\^|\^\fB4\fR\^|\^\fB5\fR\^|\^\fB6\fR) and \fBshibata-A-saturated\fR.
.hy
The latter \fBshibata-\fR ones use the new shaper coefficients
from Naoki Shibata's \fBSSRC\fR package, described at https://shibatch.org/ssrc
.SP
The filter types are distinguished by the following properties:
audibility of noise, level of (inaudible, but in some circumstances
problematic) shaped high frequency noise and processing speed
and they are available for the following sample rates:
.SP
.ne 8
.TS
center;
lI lI
l l.
Filter	Sample rates
lipshitz	44100
e- and f-weighted	48000
gesemann	44100, 48000
shibata	8000, 11025, 16000, 22050
\ 	32000, 37800, 44100, 48000
low-shibata	44100, 48000
high-shibata	44100
shibaba-A0 and A1	8000, 11025, 22050, 44100,
\ 	48000, 88200, 96000, 192000
shibata-A2	44100, 48000, 88200, 96000, 192000
shibata-A3 to A6	44100, 48000
shibata-B0 to B6	44100, 48000
shibata-A-saturated	8000, 11025, 22050
.TE
.SP
The
.B \-a
option enables a mode where dithering (and noise-shaping if applicable)
are automatically enabled only when needed.  The most likely use for
this is when applying fade in or out to an already dithered file, so
that the redithering applies only to the faded portions.  However, auto
dithering is not foolproof, so the fades should be checked carefully
for any noise modulation; if this occurs, then either redither the whole
file or use \fBtrim\fP and \fBfade\fR and concatenate the results.
.SP
The
.B \-p
option overrides the target precision in bits and can be from 1 to 24.
.SP
If the SoX global option
.B \-R
option is not given, the pseudo-random number generator used to
generate the white noise is reseeded, i.e. the generated noise
will be different on every invocation.
.SP
If the target precision is 1-bit, the \fBsdm\fR effect is applied
automatically with default settings. Invoke it manually to control its
options.
.SP
See the above section on \fBDithering\fR.
.TP
\fBdolbyb\fR [\fB\-e\fR\^|\^\fBd\fR] [\fB\-u\ \fIupsamp\fR] [\fB\-h\fR] [\fB\-t\ \fIgain\fR(1.0)] [\fB\-a\ prec\fR(-5.0)] [\fB\-f\ \fR{\fB1\fR\^|\^\fB2\fR\^|\^\fB3\fR\^|\^\fB4\fR}]
\fBdolbyb\fR is a Dolby B decoder/encoder based on dolbybcsoftwaredecode
which simulates the operation of a Dolby\ B en/decoder's electronic circuit.
.SP
By default, \fBdolbyb\fR applies Dolby\ B decoding to its input signal;
with \fB\-e\fR it does Dolby\ B encoding. \fB\-d\fR is also accepted
but only for symmetry, as it is the default mode of operation.
.SP
\fB\-u\fR sets the upsampling ratio to use in the sliding filter.
Digital filtering only works well if the sample rate is well above
the cutoff frequency of the filter. For Dolby\ B's sliding filter,
that frequency can be as high as 34kHz and this does not work well
if the sample rate is only 44.1Khz. To get around this, it upsamples
the audio to a higher rate when it passes through this filter.
By default, \fB\-u0\fR, the upsampling rate is set automatically
so that the upper sample rate is at least 200Khz;
upsampling can be switched off with \fB\-u1\fR.
.SP
If \fB\-h\fR is given, upsampling is used throughout the effect
from when the audio enters to when it leaves, not just in the sliding filter.
As \fBdolbyb\fR's up/downsampling algorithm is simple
(repeating and averaging samples) you may obtain higher quality results
by upsampling with \fBrate\fR before \fBdolbyb\ \-u1\fR and
downsampling it afterwards.
.SP
\fB\-t\fR ("threshold") adjusts the gain when the audio is fed
to the Dolby gain control circuits.
When a tape deck is encoding or decoding a magnetic tape,
it knows the signal level at the tape heads but with audio files
the maximum signal level may not accurately represent the tape's
maximum flux density (200nWb/m for cassette tapes), giving erroneous results.
The \fB\-t\fR option adjusts the volume level at which
the sliding filter reacts to overcome this.
Its default value is 1.0, which assumes that the maximum amplitude
of the signal represents the maximum recording level on tape;
higher values assume that it was recorded too quietly and values
below 1.0 are for when it was recorded too loud.
.SP
To begin with, when you have little idea of what level to use,
try a wide range of levels like 5, 10, 15 and 20.
If the result sounds muffled, the threshold is too low and
if it seems to have too much treble, the threshold is too high.
Once you know the approximate level, you can try more closely-spaced levels
and listen carefully to find the best level possible.
Logic would suggest listening to where tracks fade out, to see if
the treble increases, but this method doesn't seem to work well
and the best way seems to be to see how low the level can be set
before the results sound dull and muffled, then choose a level
a bit higher than this; you can just about hear the difference
between results that differ in threshold setting by about 2.
.SP
In decode mode, the program has to use trial and error to get the
right output sample values. \fB\-a\fR sets how accurate it
needs to be before it is considered OK. A figure of 0.0\ dB would
mean an accuracy of about 1 sample value. The default is -5.0\ dB,
which is accurate to less than one sample value.
.SP
The keymap for \fIgain\fR is \fRdolbyb.gain\fB. For example:
.XE
   play_ng -V -k D:dolbyb.gain+2 -k d:dolbyb.gain-2 in.wav
.XX
lets you adjust the Threshold Gain in \(pl/\(mi 2dB steps;
to see what the new value is as you proceed, use `\-V`.
.SP
\fB\-f\fR selects one of four types of filter to use.
The program originally simulated an analog circuit for a Dolby\ B
noise reducer. However, too much filtering in the side path was
altering the phase of the side path audio, which caused problems
when the side path was recombined with the main signal. Basically
signals don't add together very well if there is too much difference
in the phase.
To fix this, there are now 4 filter modes with hopefully less of a
phase change:
.RS
.TP
.B \-f1
is the original method.
.TP
.B \-f2
is a newer method that seems to work better than 1.
.TP
.B \-f3
is another rearrangement which in practice doesn't seem to
be any better than 1.
.TP
.B \-f4
seems to work best, hence it is the default mode.
.PP
For further detail on these parameters and advice on
digitizing and processing Dolby\ B-encoded tapes, consult
the wiki pages at https://codeberg.org/sox_ng/libdolbyb
.RE
.TP
\fBdop\fR
DSD over PCM. 1-bit DSD data is packed into 24-bit samples for
transport over non-DSD-aware links.
.TP
\fBdownsample\fR [\fIfactor\fR(2)]
Downsample the signal by an integer factor: Only the first of
each \fIfactor\fR samples is retained, the others are discarded.
.SP
No decimation filter is applied. If the input is not a properly
band-limited baseband signal, aliasing will occur. This may be
desirable, e.g., for frequency translation.
.SP
The new lower sample rate propagates forward in the effects chain but,
unless you specify the new sample rate with \fB\-r\fP before the output filename
or with a final (no-op) \fBrate\fR effect, it will be resampled back up to the
original sample rate.
.SP
For a general resampling effect with antialiasing, see \fBrate\fR.
See \fBupsample\fR.
.TP
\fBearwax\fR
This effect takes a 44.1kHz stereo signal and adds audio cues that,
when listened to on headphones, move the sound stage from inside your head
to outside and in front of you, as if listening to loudspeakers.
.SP
To see how \fBearwax\fR works, say \fBsox_ng \-h earwax\fR.
.TP
\fBecho \fIgain-in gain-out\fR <\fIdelay decay\fR>
Add echoes to the audio.
In nature, echoes are reflected sound and digital echo
effects emulate this and are often used to help fill
out the sound of a single instrument or vocal.
.SP
\fIGain-in\fR controls how much of the input signal is delivered clean
to the output,
\fIdelay\fR is the time difference in milliseconds
between the original signal and its reflection,
\fIdecay\fR is the loudness of the reflected signal and
\fIgain-out\fR is a final volume adjustment of the result.
.SP
There are keymaps for \fIgain_in\fR and \fIgain_out\fR.
.SP
There is no limit to the number of delay/decay pairs you can use
and gains and decays can be negative or greater than 1 if you wish.
.SP
\fBecho\fP extends the length of the signal by the maximum delay time.
.SP
For example,
this makes it sound as if there are twice as many instruments as are
actually playing:
.XE
   play_ng lead.aiff echo 0.8 0.88 60 0.4
.XX
If the delay is very short, it sound like a metallic robot:
music:
.XE
   play_ng lead.aiff echo 0.8 0.88 6 0.4
.XX
A longer delay sounds like an open air concert in the mountains:
.XE
   play_ng lead.aiff echo 0.8 0.9 1000 0.3
.XX
One mountain more, and:
.XE
   play_ng lead.aiff echo 0.8 0.9 1000 0.3 1800 0.25
.XX
For a flow diagram of how \fBecho\fR works,
say \fBsox_ng \-h echo\fR.
.TP
\fBechos \fIgain-in gain-out\fR <\fIdelay decay\fR>
.SP
Echos stands for `Echo in Sequel' and adds a sequence of echoes to the audio.
That is, the first echo takes the input,
the second the input and the first echo,
the third the input and the output of the second echo and so on.
A single \fBechos\fR has the same effect as a single \fBecho\fR.
Each
.I "delay decay"
pair gives the delay in milliseconds (with a minimum of one sample)
and the decay of that echo.
\fIGain-out\fR is a final volume multiplier applied to the sum of
the input\ \(mu\ \fIgain-in\fR and the delays' outputs\ \(mu
their respective decays, and there are keymaps for both.
.SP
There are keymaps for \fIgain_in\fR and \fIgain_out\fR.
.SP
\fBechos\fP extends the length of the signal by the maximum delay time.
.SP
For example:
.SP
The sample is bounced twice in symmetric echos:
.XE
   play_ng lead.aiff echos 0.8 0.7 700 0.25 700 0.3
.XX
The sample is bounced twice in asymmetric echos:
.XE
   play_ng lead.aiff echos 0.8 0.7 700 0.25 900 0.3
.XX
The sample sounds as if it were played in a garage:
.XE
   play_ng lead.aiff echos 0.8 0.7 40 0.25 63 0.3
.XX
For a flow diagram of how \fBechos\fR works,
say \fBsox_ng \-h echos\fR.
.TP
\fBequalizer \fIfrequency width\fR[\fBq\fR\^|\^\fBo\fR\^|\^\fBh\fR\^|\^\fBk\fR] \fIgain\fR
Apply a two-pole peaking equalization filter.
With this filter, the signal level at and around a selected frequency
can be increased or decreased while, unlike band-pass and band-reject
filters, the level at all other frequencies is unchanged.
.SP
\fIfrequency\fR gives the filter's central frequency in Hz,
\fIwidth\fR gives its bandwidth
and \fIgain\fR the required gain
or attenuation in dB.
Beware of
.B Clipping
when using a positive \fIgain\fR.
.SP
In order to produce complex equalization curves, this effect
can be given several times, each with a different central frequency.
.SP
The filter is described in detail in [1].
.SP
This effect supports the \fB\-\-plot\fR global option.
.TP
\fBfade\fR [\fItype\fR] \fIfade-in-length\fR [\fIstop-position(=)\fR [\fIfade-out-length\fR]]
Apply a fade effect to the beginning, end, or both of the audio.
.SP
An optional \fItype\fR can be specified to select the shape of the fade
curve:
\fBq\fR for quarter of a sine wave, \fBh\fR for half a sine
wave, \fBt\fR for linear (`triangular') slope, \fBl\fR for logarithmic,
\fBp\fR for inverted parabola and \fBs\fR for square-law.
The default is logarithmic.
.SP
A fade-in starts from the first sample and ramps the signal level from 0
to full volume over the time given as \fIfade-in-length\fR. Specify 0 if
no fade-in is wanted.
.SP
For a fade-out, the audio is truncated at
.I stop-position
and the signal level is ramped from full volume down to 0 over an
interval of \fIfade-out-length\fR before the \fIstop-position\fR. If
.I fade-out-length
is not specified, it defaults to the same value as
\fIfade-in-length\fR.
No fade-out is performed if
.I stop-position
is not specified.
If the audio length can be determined from the input file header and any
previous effects, then `\-0' (or, for historical reasons, `0') may
be specified for
.I stop-position
to indicate the usual case of a fade out that ends at the end of the input
audio stream.
.SP
See the
.B splice
effect.
.TP
\fBfir\fR [\fIcoefs-file\fR\^|\^\fIcoef\fR <\fIcoef\fR>]
Use SoX's FFT convolution engine with given Finite Impulse Response filter
coefficients.
If a single argument is given, it is the name of a file
containing the filter coefficients (white space separated; may contain
`#' comments). If the filename is `\-' or if no argument is
given, the coefficients are read from the `standard input' (stdin);
otherwise, coefficients may be given on the command line.
Examples:
.XE
   sox_ng in.au out.au fir .0195 \-.082 .234 .891 \-.145 .043

   sox_ng in.au out.au fir coefs.txt
.XX
.ne 6
with \f(CWcoefs.txt\fR containing
.XE
   # HP filter: freq=10000
     1.2311233052619888e\-01
    \-4.4777096106211783e\-01
     5.1031563346705155e\-01
    \-6.6502926320995331e\-02
.XX
This effect supports the \fB\-\-plot\fR global option.
.TP
\fBfirfit\fR [\fIknots-file\fR\^|\^<\fIfreq gain\fR>]
Use SoX's FFT convolution engine to make a filter whose frequency response
approximates a spline passing through a series of frequency/gain pairs.
If a single argument is given, it is the name of a file
containing the knots (white space separated; may contain
`#' comments).  If the given filename is `\-' or if no argument is
given, the knots are read from the `standard input' (stdin);
otherwise, knots may be given on the command line.
.SP
\fIGains\fR are in dB and the knot frequencies must be in increasing order.
.SP
Examples:
.XE
   sox_ng in.au out.au firfit 20 0 10000 -3
.XX
gives a gentle low-pass filter and
.XE
   sox_ng in.au out.au firfit knots.txt
.XX
with knots.txt containing
.XE
   # Approximate telephone response
   300  -100
   400   -10
   480     0
   2800    0
   3000  -10
   3400 -100
.XX
approximates the response of a carbon microphone telephone.
.SP
This effect supports the \fB\-\-plot\fR global option.
.TP
\fBflanger\fR [\-n\^|\^l\^|\^q] [\-s\^|\^t] [\fIdelay\fR(0) [\fIdepth\fR(2) [\fIregen\fR(0) [\fIwidth\fR(71) [\fIspeed\fR(0\*d5) [\fIshape\fR(sine)] [\fIphase\fR(25) [\fIinterp\fR(linear)]]]]]]]
Apply a flanging effect to the audio.
See [3] for a detailed description of flanging.
.SP
The parameters give the base delay and the added swept delay in milliseconds,
the percentage of regeneration (the delayed signal feedback),
\fIwidth\fR the percentage of delayed signal that is mixed with the original,
\fIspeed\fR the number of sweeps per second,
the shape of the swept wave (\fBsine\fR or \fBtriangle\fR),
the percentage of phase shift of the swept wave in multichannel flanges
(0 = 100 = the same phase on each channel) and
the type of digital delay line interpolation
(\fBnone\fR, \fBlinear\fR or \fBquadratic\fR).
.SP
\fBsine\fR, \fBtriangle\fR, \fBnone\fR, \fBlinear\fR and \fBquadratic\fR can be abbreviated.
.SP
The input and the delay's output are mixed and balanced so they
don't clip, so a \fIwidth\fR of 100 gives 50:50 mixing; to obtain only
the delayed output and none of the input, specify \fIwidth\fR as \fBinf\fR.
.SP
Despite containing a delay, \fBflanger\fR does not extend the length
of the signal so, if you also want the last dregs of the delayed output
and feedback, \fBpad\fR the signal beforehand.
.SP
\fBsine\fR, \fBtriangle\fR, \fBnone\fR, \fBlinear\fR and \fBquadratic\fR can be abbreviated
and
.BR \-s ,
.BR \-t ,
.BR \-n ,
.B \-l
and
.B \-q
are alternative ways to set the waveshape and the interpolation type
without having to specify the rest of the parameters.
.SP
For a flow diagram of how \fBflanger\fR works,
say \fBsox_ng \-h flanger\fR.
.TP
\fBgain \fR[\fB\-e\fR\^|\^\fBB\fR\^|\^\fBb\fR\^|\^\fBr\fR] [\fB\-n\fR] [\fB\-l\fR\^|\^\fBh\fR] [\fIgain-dB\fR(0)]
Apply amplification or attenuation to the audio signal or, in some
cases, to some of its channels.
Note that use of any of
.BR \-e ,
.BR \-B ,
.BR \-b ,
.B \-r
and
.B \-n
requires temporary file space to store the audio to be processed, so may
be unsuitable for use with streamed audio.
.SP
Without other options,
.I gain-dB
adjusts the signal power level by the given number of dB:
positive amplifies (beware of clipping), negative attenuates.
With other options, the
.I gain-dB
amplification or attenuation is applied after the processing due to those options.
.SP
With the
.B \-e
option, the levels of the audio channels of a multichannel file are equalized,
i.e. gain is applied to all channels other than that with the highest peak
level so that all channels attain the same peak level
(but, without also giving
.BR \-n ,
the audio is not normalized).
.SP
The
.B \-B
(balance) option is similar to
.BR \-e ,
but with
.BR \-B,
the RMS level is used instead of the peak level.
.B \-B
might be used to correct stereo imbalance caused by an imperfect record
turntable cartridge.
Note that, unlike
.BR \-e ,
.B \-B
might cause some clipping.
.SP
.B \-b
is similar to
.B \-B
but has clipping protection, i.e. if necessary to prevent clipping
whilst balancing, attenuation is applied to all channels.
In conjunction with
.BR \-n ,
.B \-B
and
.B \-b
are synonymous.
.SP
The
.B \-r
option is used in conjunction with a prior invocation of
.B gain
with the
.B \-h
option\*msee below for details.
.SP
The
.B \-n
option normalizes the audio to 0dB FSD.
It is often used in conjunction with a negative
.I gain-dB
so that the audio is normalized to a given level below 0dB.
For example,
.XE
   sox_ng in.au out.au gain \-n
.XX
normalizes to 0dB, and
.XE
   sox_ng in.au out.au gain \-n \-3
.XX
normalizes to \-3dB.
.SP
The
.B \-l
option invokes a simple limiter. For example,
.XE
   sox_ng in.au out.au gain \-l 6
.XX
applies 6dB of gain but never clips.  Note that limiting more than a
few dBs more than occasionally in a piece of audio is not recommended
as it can cause audible distortion.
See the
.B compand
effect for a more capable limiter.
.SP
The
.B \-h
option is used to apply gain to provide headroom for subsequent
processing.  For example, with
.XE
   sox_ng in.au out.au gain \-h bass +6
.XX
6dB of attenuation is applied prior to the bass boosting effect,
ensuring that it does not clip.  Of course, with \fBbass\fR, it is
obvious how much headroom is needed but, with other effects (e.g.
rate, dither), it is not always as clear.  Another advantage of using
\fBgain \-h\fR rather than an explicit attenuation is that, if the
headroom is not used by subsequent effects, it can be reclaimed with
\fBgain \-r\fR, for example:
.XE
   sox_ng in.au out.au gain \-h bass +6 rate 44100 gain \-r
.XX
The above effects chain guarantees never to clip nor amplify;
it attenuates if necessary to prevent clipping, but by only as
much as is needed to do so.
.SP
Output formatting (dithering and bit-depth reduction) also requires
headroom which cannot be reclaimed, e.g.
.XE
   sox_ng in.au out.au gain \-h bass +6 rate 44100 gain \-rh dither
.XX
Here, the second
.B gain
invocation reclaims as much of the headroom as it can from the
preceding effects but retains as much headroom as is needed for
subsequent processing.
The SoX global option
.B \-G
can be given to automatically invoke \fBgain \-h\fR and \fBgain \-r\fR.
.SP
Note that \fBsynth\fR without the \fB\-n\fR option
incorporates the functionality of \fBgain\ -h\fR.
.SP
See the
.B norm
and
.B vol
effects.
.TP
\fBhighpass\fR [\fB\-1\fR\^|\^\fB2\fR] \fIfrequency\fR [\fRwidth\fR[\fBq\fR\^|\^\fBo\fR\^|\^\fBh\fR\^|\^\fBk\fR]]
Apply a high-pass filter with 3dB point \fIfrequency\fR.
The filter can be either single-pole (with
.BR \-1 ),
or double-pole (the default, or with
.BR \-2 ).
.I width
applies only to double-pole filters;
the default is Q = 0\*d707 and gives a Butterworth response.  The filters
roll off at 6dB per pole per octave (20dB per pole per decade).  The
double-pole filters are described in detail in [1].
.SP
This effect supports the \fB\-\-plot\fR global option.
.SP
See \fBsinc\fR for filters with a steeper roll-off.
.TP
\fBhilbert\fR [\fB\-n \fItaps\fR]
Apply an odd-tap Hilbert transform filter, phase shifting the signal
by 90 degrees.
.SP
This is used in many matrix coding schemes and for analytic signal
generation.  The process is often written as a multiplication by \fIi\fR
(or \fIj\fR), the imaginary unit.
.SP
An odd-tap Hilbert transform filter has a band-pass characteristic,
attenuating the lowest and highest frequencies.  Its bandwidth can be
controlled by the number of filter taps which, by default, is chosen
for a cutoff frequency of about 75 Hz. For a cutoff frequency of about N Hz,
give the \fB\-n\fR option with the sample rate divided by N.
The number of taps can be from 3 to 1,073,741,823 but the maximum value
requires 56GB of physical RAM to complete within minutes rather than days
and 100,000,001 requires 18GB.
.SP
This effect supports the \fB\-\-plot\fR global option.
.TP
\fBladspa\fR [\fB\-l\fR] [\fB\-r\fR] \fImodule\fR [\fIplugin\fR] {\fIargument\fR}
Apply a LADSPA [5] (Linux Audio Developer's Simple Plugin API) plugin.
Despite the name, LADSPA is not Linux-specific and a wide range of
effects is available as LADSPA plugins, such as CMT [6] (the Computer
Music Toolkit) and Steve Harris's plugin collection [7]. The first
argument is the plugin module, the second the name of the plugin (a
module can contain more than one plugin) and any other arguments are
for the control ports of the plugin. Missing arguments are supplied by
default values if possible.
.SP
Normally, the number of input ports of the plugin must match the number
of input channels and the number of output ports determines the output
channel count.  However, the
.B \-r
(replicate) option allows cloning a mono plugin to handle multichannel
input.
.SP
Some plugins introduce latency which SoX may optionally compensate for.
The
.B \-l
(latency compensation) option automatically compensates for latency
as reported by the plugin via an output control port named "latency".
.SP
If it is set, the environment variable \fBLADSPA_PATH\fP is used as
the search path for plugins.
See \fBLADSPA_PATH\fR in the section \fBENVIRONMENT\fR.
.TP
\fBloudness\fR [\fIgain\fR [\fIreference\fR]]
Loudness control is similar to the
.B gain
effect but provides equalization for the human auditory system.  See
http://en.wikipedia.org/wiki/Loudness for a detailed description of
loudness.  The gain is adjusted by the given
.I gain
parameter (usually negative) and the signal equalized according to ISO
226 w.r.t. a reference level of 65dB, though an alternative
.I reference
level may be given if the original audio has been equalized at some
other level.
A default gain of \-10dB is used if a
.I gain
value is not given.
.SP
See the
.B gain
effect.
.TP
\fBlowpass\fR [\fB\-1\fR\^|\^\fB2\fR] \fIfrequency\fR [\fRwidth\fR[\fBq\fR\^|\^\fBo\fR\^|\^\fBh\fR\^|\^\fBk\fR]]
Apply a low-pass filter.
See the description of the \fBhighpass\fR effect for details.
.TP
\fBmcompand\fR \fB\(dq\fIcompand-args\fB\(dq\fR {\fIfrequency \fB\(dq\fIcompand-args\fB\(dq\fR}
.SP
The quoted \fIcompand-args\fR are as for the \fBcompand\fR effect:
.br
\fIattack1\fB,\fIdecay1\fR{\fB,\fIattack\fB,\fIdecay\fR}
.br
[\fIsoft-knee-dB\fB:\fR]\fIin-dB1\fR[\fB,\fIout-dB1\fR]{\fB,\fIin-dB\fB,\fIout-dB\fR}
.br
[\fIgain\fR [\fIinitial-volume-dB\fR [\fIdelay\fR]]]
.SP
The multi-band compander is similar to the single-band compander but the
audio is first divided into bands using Linkwitz-Riley crossover filters
and a separately specifiable compander is run on each band.  See the
\fBcompand\fR effect for the definition of its parameters.  Compand
parameters are specified between double quotes and the crossover
frequency for that band is given by \fIcrossover-freq\fR; these can be
repeated to create multiple bands.
.SP
The following examples approximate Dolby A compression and decompression,
as used for tape noise reduction in professional recording studios:
.XE
.ne 6
   # Dolby A compressor
   sox_ng in.au dolbyA.au mcompand \e
      ".1,.1 4:-56,-46,-36,-26,-26,-20,-17,-15,-9,-9" 80 \e
      ".1,.1 4:-56,-46,-36,-26,-26,-20,-17,-15,-9,-9" 3k \e
      ".1,.1 4:-56,-46,-36,-26,-26,-20,-17,-15,-9,-9" 9k \e
      ".1,.1 4:-56,-42,-36,-23,-26,-18,-17,-14,-9,-9"

.ne 6
   # Dolby A decompressor
   sox_ng dolbyA.au out.au mcompand \e
      ".1,.1 4:-46,-56,-26,-36,-20,-26,-15,-17,-9,-9" 80 \e
      ".1,.1 4:-46,-56,-26,-36,-20,-26,-15,-17,-9,-9" 3k \e
      ".1,.1 4:-46,-56,-26,-36,-20,-26,-15,-17,-9,-9" 9k \e
      ".1,.1 4:-42,-56,-23,-36,-18,-26,-14,-17,-9,-9"
.XX
Real Dolby A probably compands each channel separately
but that is left as an exercise to interested readers.
.SP
See
.B compand
for a single-band companding effect.
.TP
\fBnoiseprof\fR [\fIprofile-file\fR]
Calculate a profile of the audio for use in noise reduction.  See the
description of the \fBnoisered\fR effect for details.
.TP
\fBnoisered\fR [\fIprofile-file\fR [\fIamount\fR]]
Reduce noise in the audio signal by profiling and filtering.  This
effect is moderately effective at removing consistent background noise
such as hiss or hum.  To use it, first run SoX with the \fBnoiseprof\fR
effect on a section of audio that ideally would contain silence but in
fact contains noise\*msuch sections are typically found at the beginning
or the end of a recording.  \fBnoiseprof\fR writes a noise
profile to \fIprofile-file\fR or to stdout if no \fIprofile-file\fR or
if `\-' is given.  E.g.
.XE
   sox_ng speech.wav \-n trim 0 1.5 noiseprof speech.noise-profile
.XX
To actually remove the noise, run SoX again, this time with the \fBnoisered\fR
effect;
.B noisered
reduces noise according to a noise profile generated by
.BR noiseprof ,
from
.I profile-file
if it is given
or from stdin if no \fIprofile-file\fR or if `\-' is given.  E.g.
.XE
   sox_ng speech.wav cleaned.wav noisered speech.noise-profile 0.3
.XX
How much noise should be removed is specified by
.IR amount \*ma
number between 0 and 1 with a default of 0\*d5.  Higher numbers
remove more noise but present a greater likelihood of removing wanted
components of the audio signal.  Before replacing an original recording
with a noise-reduced version, experiment with different
.I amount
values to find the optimal one for your audio; use headphones to check
that you are happy with the results, paying particular attention to quieter
sections of the audio.
.SP
On most systems, the two stages\*mprofiling and reduction\*mcan be combined
using a pipe, e.g.
.XE
   sox_ng noisy.wav \-n trim 0 1 noiseprof | \e
      play_ng noisy.wav noisered
.XX
.TP
\fBnorm\fR [\fIdB-level(0)\fR]
Normalize the audio.
.B norm
is just an alias for \fBgain \-n\fR; see the
.B gain
effect for details.
.TP
\fBoops\fR
Out Of Phase Stereo effect.
Mixes stereo to twin mono where each mono channel contains the
difference between the left and right stereo channels.
This is sometimes known as the `karaoke' effect as it often has the effect
of removing most or all of the vocals from a recording.
It is equivalent to \fBremix 1,2i 1,2i\fR.
.SP
See also \fBcentercut\fR.
.TP
\fBoverdrive\fR [\fIgain\fR(20) [\fIcolor\fR(20)]]
Non-linear distortion.
The \fIcolor\fR parameter controls the amount of even harmonic content
in the overdriven output. Both parameters range from 0 to 100.
.SP
There are keymaps for both \fIgain\fR and \fIcolor\fR.
.TP
\fBpad\fR { \fI[%]length\fR[\fB@\fIposition(=)\fR] }
Pad the audio with silence at the beginning, at the end or at any
specified points throughout the audio.
.I length
is the amount of silence to insert and
.I position
the position in the input audio stream at which to insert it.
Any number of lengths and positions may be specified, provided that
a specified position is not less that the previous one.
.I Position
is optional for the first and last lengths specified and
if omitted correspond to the beginning and the end of the audio respectively.
For example,
.B pad 1.5 1.5
adds 1\*d5 seconds of silence at each end of the audio, whilst
.B pad 4000s@3:00
inserts 4000 samples of silence 3 minutes into the audio.
If silence is wanted only at the end of the audio, either specify the end
position or specify a zero-length pad at the start.
.SP
If a pad specification starts with with a \fB%\fR sign, the output is padded
to a multiple of \fIlength\fR at the specified position. For example,
\fBpad 0 %10\fR adds silence at the end of the audio up to the next
multiple of 10 seconds.
.SP
See
.B delay
for an effect that can add silence at the beginning of
the audio on a channel-by-channel basis.
.TP
\fBphaser\fR [\fB\-n\fR\^|\^\fBl\fR\^|\^\fBq\fR] [\fB\-s\fR\^|\^\fBt\fR] [\fIgain-in\fR(.4) \fIgain-out\fR(.74) \fIdelay\fR(3) \fIregen\fR(.4) \fIspeed\fR(.5) [\fB\-s\fR\^|\^\fBt\fR]
Add a phasing effect to the audio.
See [3] for a detailed description of phasing.
.SP
\fIdelay\fR gives the maximum delay in milliseconds from 0 to 1000,
\fIregen\fR the amount of feedback from the delay from \(mi1 to \(pl1
and \fIspeed\fR the frequency of delay-time modulation wave in Hz.
.SP
The modulation is either sinusoidal (\fB\-s\fR, the default),
which is preferable for multiple instruments, or triangular
(\fB\-t\fR) which gives single instruments a sharper phasing effect.
\fIregen\fR can be from \(mi1 to \(pl1 but should usually be less than
0\*d5 to avoid clipping and \fIgain-out\fR is the final volume adjustment
from \(mi1 to \(pl1.
.SP
The \fB\-l\fR flag makes \fBphaser\fR do linear interpolation between samples
when the offset into the delay line is not a whole number,
which is about 15% slower but much less noisy and
\fB\-q\fR does quadratic interpolation, which is about 50% slower
but even less noisy.
\fB\-n\fR explicitly asks for no interpolation, the default, fast and fuzzy.
.SP
In sox_ng, \fB\-s\fR or \fB\-t\fR can be given at the start or the end;
to be compatible with earlier versions of SoX, supply all the parameters
with one of these at the end, use \fIgain-in\fR and \fIgain-out\fR from 0 to 1,
\fIdelay\fR from 0 to 5, \fIspeed\fR from 0\*d1 to 2
and don't use interpolation.
.SP
There are keymaps for \fIgain_in\fR, \fIgain_out\fR and \fIregen\fR.
.SP
Technically, the SoX \fBphaser\fP is not a phaser; it is a flanger.
A flanger does comb filtering with equidistant spacing
(e.g. 100Hz, 200Hz, 300Hz, 400Hz, ...), while a real phaser does
comb filtering with factored spacing
(e.g. 100Hz, 200Hz, 400Hz, 800Hz, ...) that sounds more harmonic.
.SP
For example:
.XE
   play_ng snare.flac phaser 0.8 0.74 3 0.4 0.5 \-t
.XX
Gentler:
.XE
   play_ng snare.flac phaser 0.9 0.85 4 0.23 1.3 \-s
.XX
A popular sound:
.XE
   play_ng snare.flac phaser 0.89 0.85 1 0.24 2 \-t
.XX
More severe:
.XE
   play_ng snare.flac phaser 0.6 0.66 3 0.6 2 \-t
.XX
For a flow diagram of how \fBphaser\fR works,
say \fBsox_ng \-h phaser\fR.
.TP
\fBpitch \fR[\fB\-q\fR] \fIshift\fR [\fIsegment\fR [\fIsearch\fR [\fIoverlap\fR]]]
Change the audio pitch but not the tempo.
.SP
.I shift
gives the pitch shift as positive or negative `cents' (i.e. 100ths of a
semitone).
.SP
Note that raising the pitch increases the sample rate and this can
make following effects slower, in particular \fBpitch\fR or \fBtempo\fR
themselves, whose running times are proportional to the sample rate times
the overlap, all squared.
This can be compensated for by following \fBpitch\fR with
a fast \fBrate\fR effect.
.SP
.B Pitch
and
.B tempo
share the same fundamental algorithm; see the \fBtempo\fR
effect for a description of the other parameters.
.SP
See the \fBbend\fR, \fBspeed\fR and \fBtempo\fR effects.
.TP
\fBrate\fR [\fB\-q\fR\^|\^\fBl\fR\^|\^\fBm\fR\^|\^\fBg\fR\^|\^\fBh\fR\^|\^\fBe\fR\^|\^\fBv\fR\^|\^\fBu\fR] [override-options] [\fIfrequency\fR]
Change the audio sampling rate (i.e. resample the audio) to any given
.I frequency
(even non-integer if this is supported by the output file format)
using a quality level defined as follows:
.SP
.ne 10
.TS
center;
cI cI cI cI lxI
cB c c c lx.
\ 	Quality	T{
B/W
T}	Rej dB	T{
.na
Typical Use
.ad
T}
\-q	T{
quick
T}	n/a	T{
\(~=30\^@\^Fs/4
T}	T{
.na
playback on ancient hardware
.ad
T}
\-l	low	80%	100	T{
.na
playback on old hardware
.ad
T}
\-m	medium	95%	100	T{
.na
audio playback
.ad
T}
\-g	generic	95%	100	T{
16-bit
T}
\-h	high	95%	125	T{
.na
20-bit for 16-bit mastering
.ad
T}
\-e	extreme	95%	150	T{
24-bit
T}
\-v	T{
.na
very high
.ad
T}	95%	175	28-bit for 24-bit mastering
\-u	T{
ultra
T}	95%	200	32-bit
.TE
.SP
These can also be selected with \fB\-Q\ \fIn\fR with \fIn\fR from 0 to 7.
.SP
B/W (bandwidth) is the percentage of the audio frequency band that is preserved and
.I Rej dB
is the level of noise rejection.  Increasing levels of resampling
quality come at the expense of increasing amounts of time to process the
audio.  If no quality option is given, the quality level used is `high'
when processing audio and `low' when playing it.
See \fBPlaying & Recording Audio\fR above.
.SP
The `quick' algorithm uses cubic interpolation; all others use
band-limited interpolation.  By default, all algorithms have
a linear phase response; for `medium' and above,
the phase response is configurable (see below).
.SP
The
.B rate
effect is invoked automatically if SoX's \fB\-r\fR option specifies a
rate that is different to that of the input file(s).  Alternatively, if
this effect is given explicitly, then SoX's
.B \-r
option need not be given.  For example, the following two commands are
equivalent:
.XE
.ne 2
   sox_ng input.wav \-r 48k output.wav bass \-b 24
   sox_ng input.wav        output.wav bass \-b 24 rate 48k
.XX
though the second command is more flexible as it allows
.B rate
options to be given, and allows the effects to be ordered arbitrarily.
.SP
A user notes that resampling tracks and then concatenating them
is more likely to create clicks at the joints than joining them first
and resampling the result, due to edge effects.
.SP
.ne 3
.B "Override Options"
.SP
The simple quality selection described above provides settings that
satisfy the needs of the vast majority of resampling tasks.
Occasionally, however, it may be desirable to fine-tune the resampler's
filter response; for qualities `medium' and above,
this can be achieved using the override options
in the following table:
.SP
.ne 6
.TS
center;
lB lx.
\-M/\-I/\-L	Phase response=minimum/intermediate/linear
\-s	Steep filter (bandwidth=99%)
\-a	Allow aliasing/imaging above the pass band
\-b\ \fIwidth\fR	Any bandwidth % (74\-99\*d7 or 85-99\*d7 with \B\-a\fR)
\-p\ \fIphase\fR	T{
.na
Any phase response (0=minimum, 25=intermediate, 50=linear, 100=maximum)
.ad
T}
.TE
.SP
All resamplers use filters that can sometimes create `echo' (a.k.a.
`ringing') artefacts with transient signals such as those that occur
with `finger snaps' or other highly percussive sounds.  Such artefacts are
much more noticeable to the human ear if they occur before the transient
(`pre-echo') than if they occur after it (`post-echo').  Note that the
frequency of any such artefacts is related to the smaller of the
original and new sampling rates but if this is at least 44\*d1kHz,
the artefacts will lie outside the range of human hearing.
.SP
A phase response setting may be used to control the distribution of any
transient echo between
`pre' and `post': with minimum phase, there is no pre-echo but the
longest post-echo; with linear phase, pre- and post-echo are in equal
amounts (in signal terms, but not in audibility); the intermediate
phase setting attempts to find the best compromise by selecting a small
length (and level) of pre-echo and a medium-length of post-echo.
.SP
A minimum, intermediate or linear phase response is selected using the
\fB\-M\fR, \fB\-I\fR and \fB\-L\fR options;
a custom phase response can be created with the
.B \-p
option.  Note that phase responses between `linear' and `maximum'
(greater than 50) are rarely useful.
.SP
A resampler's bandwidth setting determines how much of the frequency
content of the original signal (w.r.t. the original sample rate when
upsampling or the new sample rate when downsampling) is preserved
during conversion.  The term `pass band' is used to refer to all frequencies
up to the bandwidth point (e.g. for a 44\*d1kHz sampling rate and a
resampling bandwidth of 95%, the pass band represents frequencies from
0Hz (DC) to circa 21kHz).  Increasing the resampler's bandwidth
results in a slower conversion and can increase transient echo
artefacts (and vice versa).
.SP
The
.B \-s
`steep filter' option changes the resampling bandwidth from the default of 95%
(based on the 3dB point) to 99%.  The
.B \-b
option allows the bandwidth to be set to any value in the range
74\-99\*d7% but bandwidth values greater than 99% are not
recommended for normal use as they can cause excessive transient echo.
.SP
If the
.B \-a
option is given, aliasing/imaging above the pass band is allowed.  For
example, with 44\*d1kHz sampling rate and a
resampling bandwidth of 95%, this means that frequency content above
21kHz can be distorted. However, since this is above the pass band (i.e.
above the highest frequency of interest/audibility), this may not be a
problem.  The benefits of allowing aliasing/imaging are reduced processing time
and reduced (by almost half) transient echo artefacts.
.SP
The
.B \-d
option sets the bit-accuracy in the range 15 to 33, or
.B \-R
sets the bit-accuracy to obtain rejection of a specified number of dB.
.SP
Examples:
.XE
   sox_ng input.wav \-b 16 output.wav rate \-s \-a 44100 dither \-s
.XX
is default (high) quality resampling with overrides for a steep filter,
to allow aliasing, at a 44\*d1kHz sample rate and noise-shaped dithering
to a 16-bit WAV file.
.XE
   sox_ng input.wav \-b 24 output.aiff rate \-v \-I \-b 90 48k
.XX
is very high quality resampling with overrides for an intermediate phase,
a bandwidth of 90%, at a 48k sampling rate and storing the output
to a 24-bit AIFF file.
.SP
.ne 3
.B "Advanced Options"
.SP
The
.B \-i
option forces the use of a particular interpolator coefficient from -1 to 2.
.SP
The
.B \-c
option tries to limit the number of coefficients to a number of kilobytes;
its argument can be from 100 up.
.SP
The
.B \-B
option sets the percentage of the pass-band to preserve, from 53 to 95.
.SP
The
.B \-A
option sets the percentage of the bandwidth without aliasing, from 85 to 100.
.SP
.B \-f
sets zero pass-band roll-off instead of 0.01dB for -Q 0-2.
.SP
.B \-n
disables internal small-integer optimizations and
.SP
.B \-t
increases the irrational ratio accuracy.
.TP
\fBremix\fR [\fB\-a\fR\^|\^\fBm\fR] [\fB\-p\fR] <\fIout-spec\fR>
\fIout-spec\fR	= \fB0\fR | \fIin-spec\fR{\fB,\fIin-spec\fR}
.br
\fIin-spec\fR	= [\fIin-chan\fR]\^[\fB\-\fR[\fIin-chan2\fR]]\^[\fIvol-spec\fR]
.br
\fIvol-spec\fR	= \fBp\fR\^|\^\fBi\fR\^|\^\fBv\^\fR[\fIvolume\fR]
.br
.SP
Select and mix input audio channels into output audio channels.  Each output
channel is specified in turn by a given \fIout-spec\fR which is a list of
contributing input channels and volume specifications.
.SP
Note that this effect operates on the audio channels
within the SoX effects processing chain; it should not be confused with the
.B \-m
global option, where multiple files
are mix-combined before entering the effects chain.
.SP
An
.I out-spec
contains comma-separated input channel numbers and hyphen-delimited
channel number ranges; alternatively,
.B 0
may be given to create a silent output channel.  For example,
.XE
   sox_ng input.wav output.wav remix 6 7 8 0
.XX
creates an output file with four channels, where channels 1, 2, and 3 are
copies of channels 6, 7, and 8 in the input file, and channel 4 is silent.
Whereas
.XE
   sox_ng input.wav output.wav remix 1\-3,7 3
.XX
creates a (somewhat bizarre) stereo output file where the left channel
is a mix-down of input channels 1, 2, 3 and 7 and the right channel is
a copy of input channel 3.
.SP
Where a range of channels is specified, the channel numbers to the left and
right of the hyphen are optional and default to 1 and to the number of input
channels respectively. Thus
.XE
   sox_ng input.wav output.wav remix \-
.XX
performs a mix-down of all input channels to mono.
.SP
By default, where an output channel is mixed from multiple input
channels, each input channel is scaled by a factor of \(S1/\s-2n\s+2.
Custom mixing volumes can be set by following a given input channel or range
of input channels with a \fIvol-spec\fR (volume specification)
which is one of the letters \fBp\fR, \fBi\fR, or \fBv\fR,
followed by a volume number, the meaning of which depends on the given
letter:
.SP
.ne 7
.TS
center;
lI lI lxI
cb l lx.
Letter	Volume number	Notes
p	power adjust in dB	0 = no change
i	power adjust in dB	T{
.na
As for \fBp\fR but invert the audio
.ad
T}
v	voltage multiplier	T{
.na
1 = no change;
0\*d5\ \(~=\ 6dB\ attenuation;
2\ \(~=\ 6dB\ gain;
\-1\ =\ invert
.ad
T}
.TE
.SP
If an \fIout-spec\fR includes at least one \fIvol-spec\fR then, by default,
\(S1/\s-2n\s+2 scaling is not applied to any other channels in the
same \fIout-spec\fR (though maybe in other \fIout-spec\fRs) though
the \fB\-a\fR (automatic)
option can be given to retain the automatic scaling in this case.
For example,
.XE
   sox_ng input.wav output.wav remix 1,2 3,4v0.8
.XX
results in channel level multipliers of 0\*d5,0\*d5 and 1,0\*d8, whereas
.XE
   sox_ng input.wav output.wav remix \-a 1,2 3,4v0.8
.XX
results in channel level multipliers of 0\*d5,0\*d5 and 0\*d5,0\*d8.
.SP
The \fB\-m\fR (manual) option disables all automatic volume adjustments, so
.XE
   sox_ng input.wav output.wav remix \-m 1,2 3,4v0.8
.XX
results in channel level multipliers of 1,1 and 1,0\*d8.
.SP
The volume number is optional and omitting it corresponds to no volume
change; however, the only case in which this is useful is in conjunction
with
.BR i .
For example, if
.I input.wav
is stereo, then
.XE
   sox_ng input.wav output.wav remix 1,2i
.XX
is a mono equivalent of the
.B oops
effect and
.XE
   play file.mp3 remix 1,2 1i,2i
.XX
lets you get twice as much power from a mono speaker connected between
the left and right poles than you would from two connected the usual way,
(but mind it doesn't blow the amplifier, as it draws twice as much current).
.SP
If the \fB\-p\fR option is given, any automatic \(S1/\s-2n\s+2 scaling
is replaced by \(S1/\s-2\(srn\s+2 (`power') scaling; this gives a louder mix
but one that may occasionally clip.
.SP
One use of the
.B remix
effect is to split an audio file into a set of files, each containing
one of the constituent channels in order to perform subsequent
processing on individual audio channels.  When more than a few
channels are involved, a script such as the following is useful:
.ne 8
.XE
#! /bin/sh
chans=\`soxi_ng \-c "$1"\`
while [ $chans \-ge 1 ]; do
   chans0=\`printf %02i $chans\`   # 2 digits hence up to 99 chans
   out=\`echo "$1" | sed "s/\\(.*\\)\\.\\(.*\\)/\\1\-$chans0.\\2/"\`
   sox_ng "$1" "$out" remix $chans
   chans=\`expr $chans \- 1\`
done
.XX
If a file
.I input.wav
containing six audio channels were given, the script would produce six
output files:
.IR input-01.wav ,
\fIinput-02.wav\fR, ...,
.IR input-06.wav .
.SP
See the \fBswap\fR effect.
.TP
\fBrepeat\fR [\fIcount\fR(1)\^|\^\fB\-\fR]
Repeat the entire audio \fIcount\fR times, or once if \fIcount\fR is not given.
The special value \fB\-\fR requests infinite repetition.
It requires temporary file space to store the audio to be repeated.
Note that repeating once yields two copies: the original audio and the
repeated audio.
.TP
\fBreverb\fR [\fB\-w\fR] [\fIreverberance\fR(50%) [\fIHF-damping\fR(50%) [\fIroom-scale\fR(100%)
[\fIstereo-depth\fR(100%) [\fIpre-delay\fR(0ms) [\fIwet-gain\fR(0dB)]]]]]]
.SP
Add reverberation to the audio using the `freeverb' algorithm.  A
reverberation effect is sometimes desirable for concert halls that are too
small or contain so many people that the hall's natural reverberance is
diminished.  Applying a small amount of stereo reverb to a dry mono signal
usually makes it sound more natural.  See [3] for a detailed description
of reverberation.
.SP
This effect increases the volume of the audio and continues to reverberate
after the input finishes so, to prevent clipping and
keep the audible part of the final reverberation,
a typical invocation might be:
.XE
   play_ng dry.au gain \-3 pad 0 1 reverb
.XX
The
.B \-w
option can be given to select only the `wet' signal, thus allowing it to be
processed further, independently of the `dry' signal.  E.g.
.XE
   play_ng \-m in.au "|sox_ng in.au \-p reverse reverb \-w reverse"
.XX
for a reverse reverb effect.
.TP
\fBreverse\fR
Reverse the audio completely.
Requires temporary file space to store the audio to be reversed.
.TP
\fBriaa\fR
Apply RIAA vinyl playback equalization.
The sampling rate must be 44\*d1, 48, 88\*d2, 96 or 192kHz.
.SP
This effect supports the \fB\-\-plot\fR global option.
.TP
\fBsaturation\fR [\fItype\fR [\fIblend\fR [\fIoffset\fR [\fIdrive\^|\^color\^|\^threshold\fR]]]]
Add saturation, which can produce effects ranging from subtle warmth to
crunchy fuzz. The \fItype\fR parameter selects the saturation type:
\fBtanh\fR (the default), \fBsqrt\fR or \fBdiode\fR.
.SP
For all types, the \fIblend\fR parameter (default 1) controls the mixture
of wet and dry signals in the output, with 1 being fully wet. The
\fIoffset\fR parameter (default 0) adds a DC offset to the input to produce
asymmetric distortion. The offset is removed from the output, so that a zero
input level produces a zero output level, but when the input is non-zero the
output waveform is likely to be asymmetric.
.SP
The \fBtanh\fR saturation type uses the hyperbolic tangent function to
apply soft clipping. The \fIdrive\fR parameter (default 1) controls the
input gain and thus the amount of distortion.
.SP
The \fBsqrt\fR saturation type uses a mixture of two functions:
\fIx*sqrt(|x|)\fR and \fIsgn(x)*sqrt(|x|)\fR, which give different
tonal qualities to the output. The \fIcolor\fR parameter (default 0.5)
controls the mixture of these functions, with 0 being purely
\fIx*sqrt(|x|)\fR and 1 being purely \fIsgn(x)*sqrt(|x|)\fR.
.SP
The \fBdiode\fR saturation type models the effect of using a pair of
diodes to clip the signal when it exceeds a \fIthreshold\fR
(default 0.5). The \fIblend\fR parameter, by mixing the wet and dry
signals, effectively controls the amount of attenuation that occurs above
the threshold, from no attenuation when \fIblend\fR is 0 to complete
attenuation (hard clipping) when \fIblend\fR is 1.
.SP
There are keymaps for \fIblend\fR, \fIoffset\fR, \fIdrive\fR, \fIcolor\fR
and \fIthreshold\fR.
.SP
See the \fBoverdrive\fR effect for another kind of non-linear distortion.
When the \fIoffset\fR parameter is used to produce asymmetric distortion,
the \fBhighpass\fR effect can be used to rebalance the waveform's positive
and negative amplitude.
.TP
\fBsdm\fR [\fB\-f \fIfilter\fR] [\fB\-t \fIorder\fR] [\fB\-n \fInum\fR] [\fB-l \fIlatency\fR]
Apply a 1-bit sigma-delta modulator producing DSD output.  The input
should be previously upsampled, e.g. with the \fBrate\fR effect, to a
high rate, 2\*d8224MHz for DSD64.  The \fB\-f\fR option selects the
noise-shaping filter from the following list where the number indicates
the order of the filter:
.SP
.ne 5
.TS
center;
l l.
clans-4	sdm-4
clans-5	sdm-5
clans-6	sdm-6
clans-7	sdm-7
clans-8	sdm-8
.TE
.TP
\ 
The noise filter may be combined with a partial trellis/viterbi search
by supplying the following options:
.RS
.IP "\fB\-t \fIorder\fR"
Trellis order, max 32.
.IP "\fB\-n \fInum\fR"
Number of paths to consider, max 32.
.IP "\fB\-l \fIlatency\fR"
Output latency, max 2048.
.RE
.TP
\ 
The result of using these parameters is hard to predict and can include
high noise levels or instability.  Caution is advised.
.TP
\fBsilence \fR[\fB\-l\fR] \fIabove-periods\fR [\fIduration threshold\fR[\fBd\fR\^|\^\fB%\fR]]
[\fIbelow-periods duration threshold\fR[\fBd\fR\^|\^\fB%\fR]]
.SP
Removes silence from the beginning, middle or end of the audio,
where `silence' is determined by a specified threshold.
.SP
The \fIabove-periods\fR value is used to indicate whether audio should be
trimmed at the beginning of the audio. A value of zero indicates that no
silence should be trimmed from the beginning in which case
\fIduration\fR and \fIthreshold\fR are omitted.
When a
non-zero \fIabove-periods\fR is specified,
you must also specify a \fIduration\fR and \fIthreshold\fR
and it trims audio until it finds non-silence.
It will normally be 1 when trimming silence from the beginning of the audio,
but it can be increased to higher values to trim all audio up to
the Nth non-silence period.
For example, if you have an audio file with two songs that
each contains 2 seconds of silence before the song, you could specify
an \fIabove-period\fR of 2 to strip out both silences and the first song.
.SP
\fIduration\fR indicates the
amount of time for which non-silence must be detected before it stops
trimming the silence before it.
By increasing \fIduration\fR, short bursts of quiet noise
can be treated as silence and trimmed off.
\fIduration\fR has the peculiarity that a bare
number is interpreted as a sample count, not as a number of seconds.
To specify seconds, either use the \fBt\fR suffix (as in \fB2t\fR),
a decimal point (as in \fB2.0\fR) or specify minutes too (as in \fB0:02\fR).
.SP
\fIthreshold\fR indicates the maximum sample value in any channel
is considered silence. For digital audio, a value of 0 may be fine but
for audio recorded from analog you may wish to increase the value
to include background noise.
\fIthreshold\fR numbers may be suffixed with
.B d
to indicate that the value is in decibels or
.B %
to indicate a percentage of the maximum possible sample value.
By default, it is in percent.
.SP
To trim silence from the end of the audio, specify
a \fIbelow-periods\fR count, which means
to remove all audio after the last onset of silence is detected.
Normally, this will be 1 but it can
be increased to leave shorter periods of silence
and the audio that follows them intact.
For example, if you have a track with 1 second of silence in the middle
and 1 second at the end, you could set \fIbelow-period\fR to 2
to leave the middle silence and what follows it
and remove from the final silence on.
.SP
When \fIbelow-periods\fR is given, its \fIduration\fR specifies
the length of silence that must exist before audio is not copied any more.
By specifying a higher \fIduration\fR, shorter silences that are wanted
can be left in the audio.
For example, if you have a song with 1 second of silence in the middle
and 2 seconds of silence at the end, a \fIduration\fR of 2
could be used to skip over the middle silence and trim the end
instead of starting trimming from half way through.
.SP
Unfortunately, the length of the silence at the end has to be longer
than any preceding silence for this to work
so you must know the length of the silence at the end.
.SP
A more reliable way to trim silence from the end is to use
the \fBsilence\fR effect in combination with the \fBreverse\fR effect.
By first reversing the audio, you can use the \fIabove-periods\fR
to trim from what looks like the front of the file,
then reverse it again to get back to normal.
.SP
To remove silence from the middle of a file,
give a negative \fIbelow-periods\fR.
This value is then treated as positive value and is also used to indicate that the
effect should restart processing as specified by the
\fIabove-periods\fR, making it suitable for removing periods of
silence in the middle of the audio.
.SP
The
.B \-l
option indicates that \fIbelow-periods\fR' \fIduration\fR of `silent' audio
should be left intact at the beginning of each period of silence,
for example, if you want to remove long pauses between words
but do not want to remove the pauses completely.
.SP
The following example shows how this effect can be used to make a recording
that does not contain the silence that usually occurs between
pressing the record button and the start of the performance:
.XE
   rec_ng \fIparameters filename other-effects\fR silence 1 5 2%
.XX
This example should remove the start of the recording until
there's a period of non-silence longer than 0.2s and louder than 0.1%,
then start searching for a silence that's longer than 1s and quieter than 3%
and remove it if found, leaving the first 1s of it in place,
then start copying again until a silence is found
that's longer than 1s and quieter than 3%, trim that to 1s and so on.
.XE
   sox_ng in.au out.au silence -l 1 0.2 0.1% -1 1.0 3%
.XX
.TP
\fBsinc\fR [\fB\-a\fI att\fR\^|\^\fB\-b\fI beta\fR] [\fB\-p\fI phase\fR\^|\^\fB\-M\fR\^|\^\fB\-I\fR\^|\^\fB\-L\fR] \:[\fB\-t\fI tbw\fR\^|\^\fB\-n\fI taps\fR]
[\fIfreqHP\fR]\:[\fB\-\fIfreqLP\fR [\fB\-t\fR tbw\^|\^\fB\-n\fR taps]] [\fB\-r\fR]] [\fB\-d\fR]]
.SP
Apply a kaiser-windowed low-pass, high-pass, band-pass or band-reject filter
to the signal.
The \fIfreqHP\fR and \fIfreqLP\fR parameters give the frequencies of the
6dB points of a high-pass and low-pass filter that may be invoked
individually or together.  If both are
given, \fIfreqHP\fR less than \fIfreqLP\fR creates a band-pass filter and
\fIfreqHP\fR greater than \fIfreqLP\fR creates a band-reject filter.
For example, the invocations
.XE
   sinc 3k
   sinc -4k
   sinc 3k-4k
   sinc 4k-3k
.XX
create a high-pass, low-pass, band-pass and band-reject filter
respectively.
.SP
The default stop band attenuation of 120dB can be overridden with
\fB\-a\fR; alternatively, the kaiser window's `beta' parameter can be
given directly with \fB\-b\fR.
.SP
The default transition bandwidth of 5% of the total band can be
overridden with \fB\-t\fR (and \fItbw\fR in Hertz); alternatively, the
number of filter taps can be given directly with \fB\-n\fR and is
limited to the range of 11\-1,073,741,823 though the maximum requires
56GB of physical RAM if it is to complete in minutes rather than days,
while 100,000,000 requires 8GB.
.SP
If both \fIfreqHP\fR and \fIfreqLP\fR are given, a \fB\-t\fR or
\fB\-n\fR option given to the left of the frequencies applies to both
frequencies; one of these options given to the right of the frequencies
applies only to \fIfreqLP\fR.
.SP
The
.BR \-p ,
.BR \-M ,
.B \-I
and
.B \-L
options control the filter's phase response; see the \fBrate\fR effect
for details.
.SP
The
.B \-r
option controls whether the filter should round the number of taps to the closest integer
instead of truncating it.
.SP
The
.B \-d
option specifies that, if a low-pass filter is being created and
the cutoff frequency is at or above the Nyquist frequency,
the \fBsinc\fP effect should be deleted from the effects chain
instead of failing.
.SP
This effect supports the \fB\-\-plot\fR global option.
.TP
\fBsoftvol\fR [\fIvolume\fR(1.0) [\fIdouble-time\fR(0) [\fIheadroom\fR(0)]]]
The soft volume effect applies a simple multiplier to the audio
ensuring that it does not clip. When a sample would have clipped
the volume multiplier is automatically reduced to compensate.
.SP
It is a simple compander with the advantages of running fast,
having no pre- or post-echo and reacting on the crests of the wave,
so its volume-reduction glitches don't add audible noise.
.SP
\fIvolume\fR sets the initial volume multiplier;
the default of 1.0 means no change.
.SP
\fIdouble-time\fR says that the volume should slowly increase at a rate
that makes it double every \fIdouble-time\fR seconds.
A good value for usual music is 10 and the default value of 0
says that the volume should not increase automatically.
.SP
\fIheadroom\fR is in dB and limits the loudest amplitude to
less than the 32-bit maximum.
This may be necessary when the final bit-depth reduction and/or dithering
make it clip.
A value of 0.1 is sufficient to protect down to a bit-depth of 8 with dithering.
.SP
There are keymaps for \fIvolume\fR, \fIdouble_time\fR and \fIheadroom\fR.
.SP
When playing sound in interactive mode, the `\fBv\fR' and `\fBV\fR' keys
reduce and increase the volume if there is a \fBsoftvol\fR
in the effects chain. If there are more than one, which one it adjusts
is probably random.
.TP
\fBspectrogram \fR[\fIoptions\fR]
Create a spectrogram of the audio. The audio is passed unmodified
through the SoX processing chain.  This effect is optional\*mtype
\fBsox_ng \-\-help\fR and check the list of supported effects to see if
it has been included.
.SP
The spectrogram is rendered in a Portable Network Graphic (PNG) file
and shows time in the X axis, frequency in the Y axis and audio
signal magnitude in the Z axis, represented by the
color (or optionally the intensity) of the pixels in the X-Y plane.
If the audio signal contains multiple channels, these are shown
from top to bottom starting from channel 1, which is the left channel
for stereo audio.
.SP
For example, if `my.wav' is a stereo file, then
.XE
   sox_ng my.wav \-n spectrogram
.XX
creates a spectrogram of the entire file in the file
`spectrogram.png'.  More often though, analysis of a smaller portion
of the audio is required; e.g. with
.XE
   sox_ng my.wav \-n remix 2 trim 20 30 spectrogram
.XX
the spectrogram shows information only from the second (right)
channel of thirty seconds of audio starting from twenty seconds
in.  To analyze a small portion of the frequency domain, the
.B rate
effect may be used, e.g.
.XE
   sox_ng my.wav \-n rate 6k spectrogram
.XX
allows detailed analysis of frequencies up to 3kHz (half the sampling
rate) i.e. where the human auditory system is most sensitive.
See also the \fR\-R\fR option below. With
.XE
   sox_ng my.wav \-n trim 0 10 spectrogram \-x 600 \-y 200 \-z 100
.XX
the given options control the size of the spectrogram's X, Y & Z axes
(in this case, the spectrogram area of the produced image will be 600
by 200 pixels in size and the Z axis range will be 100 dB).  Note that
the produced image includes axes, legends etc. and will be larger than the
specified spectrogram size unless the \fB\-r\fR option is given:
if each spectrogram is \fIx\fR\ \(mu\ \fIy\fR and there are \fIc\fR channels,
the image will be
\fIx\fR\ \(pl\ 144 by (\fIy\fR\ \(mu\ \fIc\fR)\ \(pl\ 78,
plus \fIc\fR\ \(mi\ 1 if \fB\-a\fR was not given,
and 20 pixels higher than this if you gave \fB\-t\fR\ \fITitle\fR.
A raw spectrogram will be \fIx\fR\ by\ \fIy\fR\ \(mu\ \fIc\fR.
.SP
In this example
.XE
   sox_ng \-n \-n synth 6 tri 10k:14k spectrogram \-z 100 \-w kaiser
.XX
an analysis window with high dynamic range is selected to best
display the spectrogram of a swept triangular wave.  For a similar
example, append the following to the `chime' command in the
description of the
.B delay
effect (above):
.XE
   rate 2k spectrogram \-X 200 \-Z \-10 \-w kaiser
.XX
Options are also available to control the appearance (color set,
brightness, contrast etc.) and filename of the spectrogram; e.g. with
.XE
   sox_ng my.wav \-n spectrogram \-m \-l \-o print.png
.XX
a spectrogram is created suitable for printing on a black and white
printer.
.SP
.B Options
.RS
.IP \fB\-x\ \fInum\fR
Change the (maximum) width (X axis) of the spectrogram from its default
value of 800 pixels to a given number between 100 and a million.
See \fB\-X\fR and \fB\-d\fR.
.IP \fB\-X\ \fInum\fR
X axis pixels per second; the default is auto-calculated to fit the
audio to the X axis size if its duration is known or given with \fR\-d\fR,
or 100 otherwise.
If given without a \fB\-x\fR option when the length of the audio is known,
this option determines the width of the spectrogram;
otherwise, it affects the duration of the spectrogram.
.I num
can be from 1 (low time resolution) to 5000 (high time resolution)
and need not be an integer.  SoX
may make a slight adjustment to the given number for processing
quantization reasons; if so, SoX reports the actual number used
(viewable when the SoX global option
.B \-V
is in effect).
.SP
Note that the number of sample frames per pixel column is an integer,
so graphs whose number of samples is not a multiple of the number of columns
may render shorter than the file or with funny time legends.
.IP \fB\-y\ \fInum\fR
Sets the size of the Y axis per channel in pixels; this is the number of
frequency `bins' used in the Fourier analysis that produces the spectrogram.
By default the Y axis size is chosen automatically, depending on the
\fB-Y\fR height and the number of channels, with a minimum of 64.
.SP
The DFT size is set to 2\ \(mu\ (\fInum\fR\ \(mi\ 1) and if SoX was compiled
with FFTW, sizes of
2^a\ \(mu\ 3^b\ \(mu\ 5^c\ \(mu\ 7^d\ \(mu\ 11^e\ \(mu\ 13^f
where e\ \(pl\ f\ <\ 2 are said to be fastest. If it wasn't,
anything other than powers of two is a hundred of times slower
in which case heights of a power of two plus one will be faster.
.IP \fB\-Y\ \fInum\fR
Sets the total height of the spectrogram(s).  The default value
is 550 pixels and the maximum is a million.
If \fInum\fR is not an exact multiple of the number of channels with \fB\-r\fR,
the actual total height of the spectrogram area will be a few pixel rows less.
For non-raw spectrograms instead, the height of the graph area will be
slightly less for the same reason, slightly more for the single-pixel row
between adjacent channel graphs if \fB\-a\fR wasn't given, and the overall
height of the image will be greater by the time axes (28) and by the
title (20) if present.
.IP \fB\-z\ \fInum\fR
Z axis (color) range in dB, default 120.  This sets the dynamic range
of the spectrogram to be \-\fInum\fR\ dBFS to 0\ dBFS.
.I Num
may range from 20 to 180.  Decreasing dynamic range effectively
increases the contrast of the spectrogram display and vice versa.
.IP \fB\-Z\ \fInum\fR
Sets the upper limit of the Z axis in dBFS.
A negative
.I num
effectively increases the brightness of the spectrogram display
and vice versa.
.IP \fB\-n\fR
Normalizes the upper limit of the Z axis so that the loudest pixels
are shown using the brightest color in the palette\*ma kind of
automatic \fB\-Z\fR flag.
.IP \fB\-q\ \fInum\fR
Sets the Z axis quantization, i.e. the number of different colors (or
intensities) in which to render Z axis
values.  A small number (e.g. 4) gives a poster-like effect making
it easier to discern magnitude bands of similar level and
results in a smaller PNG file.  The number given specifies the number of
colors to use in the Z axis range; two colors are reserved to
represent out-of-range values.
.IP \fB\-w\ \fIname\fR
Select a window function: \fBHann\fR (the default), \fBHamming\fR, \fBBartlett\fR, \fBRectangular\fR, \fBKaiser\fR or \fBDolph\fR.
The spectrogram is produced using the Discrete Fourier Transform (DFT)
algorithm and a significant parameter of this algorithm is the choice of
window function.  By default, SoX uses the Hann window, which has good
all-round properties for frequency resolution and dynamic range.  For better
frequency resolution but lower dynamic range, select a Hamming window;
for higher dynamic range but poorer frequency resolution, select a
Dolph window.
.IP \fB\-W\ \fInum\fR
Window adjustment parameter.  This can be used to make small
adjustments to the Kaiser and Dolph windows.  A positive number (up to
ten) increases its dynamic range, a negative number decreases it.
.IP \fB\-s\fR
Allow slack overlapping of DFT windows.
This can, in some cases, increase image sharpness and give greater adherence
to the
.B \-x
value but at the expense of a little spectral loss.
.IP \fB\-a\fR
Suppress the display of the axis lines.  This is sometimes useful in
helping to discern artefacts at the spectrogram edges.
.IP \fB\-r\fR
Raw spectrogram: suppress the display of axes and legends.
.IP \fB\-g\fR
When there are many seconds, show them as MM:SS or HH:MM:SS.
.IP \fB\-m\fR
Creates a monochrome spectrogram (the default is color).
.IP \fB\-h\fR
Selects a high-color palette which is less visually pleasing than the default
color palette but it may make it easier to differentiate different levels.
If this option is used in conjunction with
.BR \-m ,
the result is hybrid monochrome/color palette.
.IP \fB\-p\ \fInum\fR
Permute the colors in a color or hybrid palette.
The
.I num
parameter, from 1 (the default) to 6, selects the permutation.
.IP \fB\-l\fR
Creates a `printer-friendly' spectrogram with a light background (the
default has a dark background).
.IP \fB\-A\fR
Selects an alternative, fixed color set. This is provided only for
compatibility with spectrograms produced by another package.  It should
not normally be used as it has some problems, not least, a lack of
differentiation at the bottom end which results in masking of low-level
artefacts.
.IP \fB\-t\ \fItext\fR
Set the image title, the text to display above the spectrogram.
If you need it to be `chorus' or some other effect's name,
surround it by spaces inside double quotes.
.IP \fB\-c\ \fItext\fR
Set (or clear) the image comment, the text to display below and to the
left of the spectrogram.
.IP \fB\-o\ \fIfile\fR
The name of the spectrogram output PNG file, default `spectrogram.png'.
If `-' is given, the spectrogram is sent to the `standard output'
(stdout).
.IP \fB\-L\fR
Plot the frequency on a logarithmic axis.
.IP \fB\-R\ \fIL\fR:\fIH\fR
Specify the frequency range (from \fIL\fR to \fIH\fR).
.SP
By default, the lowest frequency is 0Hz for a linear graph or
1Hz for a logarithmic graph and the highest is the Nyquist frequency;
if a single frequency is given with no colon, it sets the low frequency.
.IP \fB\-i\fR
Interpolate vertically: where there are more output pixels than
frequency bins, use a weighted average of the bins above and below
the pixel's frequency and where there are more frequency bins than
pixels, average the bins that fall in this pixel row.
.RE
.TP
\ 
.B Advanced Options
.br
In order to process a smaller section of audio without affecting other
effects or the output signal (unlike when the
.B trim
effect is used), the following options may be used:
.RS
.IP \fB\-d\ \fIduration\fR
This option sets the X axis resolution such that audio with the given
.I duration
(a time specification) fits the selected (or default) X axis width.
It defaults, if the audio length is known, to the audio length minus
the start time.
For
example,
.XE
   sox_ng input.mp3 output.wav \-n spectrogram \-d 1:00 stats
.XX
creates a spectrogram showing the first minute of the audio, while
the
.B stats
effect is applied to the entire audio signal.
.SP
See
.B \-X
for an alternative way of setting the X axis resolution.
.IP \fB\-S\ \fIposition(=)\fR
Start the spectrogram at the given point in the audio stream.  For
example
.XE
   sox_ng input.aiff output.wav spectrogram \-S 1:00
.XX
creates a spectrogram showing all but the first minute of the audio
(the output file, however, receives the entire audio stream).
.RE
.TP
\ 
For the ability to perform off-line processing of spectral data, see
.BR "stat \-freq".
.TP
\fBspeed \fIfactor\fR[\fBc\fR]
Adjust the audio speed (pitch and tempo together).  \fIfactor\fR
is either the ratio of the new speed to the old speed (greater
than 1 speeds it up, less than 1 slows it down) or, if the
letter
\fBc\fR is appended, it's the number of cents (100ths of a semitone)
by which the pitch (and tempo) should be adjusted: greater than 0
increases, less than 0 decreases.
.SP
Technically, the speed effect only changes the sample rate information,
leaving the samples themselves untouched. The \fBrate\fR effect is invoked
automatically to resample to the output sample rate, using its default
quality/speed.  For higher quality or higher speed
resampling, in addition to the \fBspeed\fR effect, specify
the \fBrate\fR effect with the desired quality option.
.SP
See the \fBbend\fR, \fBpitch\fR
and
.B tempo
effects.
.TP
\fBspeexdsp \fR[\fB\-agc\fR [\fItarget_level\fR(100)]] [\fB\-denoise\fR [\fImax_db\fR(15)]] [\fB\-dereverb\fR]
[\fB\-fps\fR \fIframes_per_second\fR(20)] [\fB\-spf\fR \fIsamples_per_frame\fR]
.SP
Use the Speex DSP library to improve perceived sound quality.
.SP
If no options are specified, the \fB\-agc\fR and \fB\-denoise\fR features are enabled.
.RS
.IP \fB\-agc\ \fR[\fItarget_level\fR]
Enable automatic gain control and optionally specify
a target volume level from 1 to 100.
.IP \fB\-denoise\ \fR[\fImax_db\fR]
Enable noise reduction and optionally specify the maximum attenuation
from 1 to 100.
.IP \fB\-dereverb\fR
Enable reverb reduction.
.IP \fB\-fps\ \fIframes_per_second\fR
Specify the number of frames per second from 1-100.
.IP \fB\-spf\ \fIsamples_per_frame\fR
Specify the number of samples per frame.
The default is derived from the \fB\-fps\fR setting
so that frames abut but do not overlap.
.RE
.TP
\fBsplice \fR [\fB\-h\fR\^|\^\fBt\fR\^|\^\fBq\fR] {\fIposition(=)\fR[\fB,\fIexcess\fR[\fB,\fIleeway\fR]]}
Splice audio sections together.  This effect provides two things over
simple audio concatenation: a (usually short) cross-fade is applied at
the join and a wave similarity comparison is made to help determine the
best place at which to make the join.
.SP
One of the options
.BR \-h ,
.BR \-t ,
or
.B \-q
may be given to select the fade envelope as half cosine wave (the default),
triangular (a.k.a. linear), or quarter cosine wave (e.g. for a cross-fade of
correlated audio).
.SP
.ne 4
.TS
center;
cI lI lI lI
cB l l l.
\ 	Audio	Fade level	Transitions
\-h	correlated	constant gain	smooth
\-t	correlated	constant gain	abrupt
\-q	uncorrelated	constant power	smooth
.TE
.SP
To perform a splice, first use the
.B trim
effect to select the audio sections to be joined together.  As when
performing a tape splice, the end of the section to be spliced onto
should be trimmed with a small
.I excess
(default 0\*d005 seconds) after the ideal joining point.  The
beginning of the audio section to splice on should be trimmed with the
same
.IR excess
before the ideal joining point plus an additional
.I leeway
(default 0\*d005 seconds).
SoX should then be invoked with the two
audio sections as input files and the
.B splice
effect given with the position at which to perform the splice\*mthis is
length of the first audio section (including the excess).
.SP
The following diagram uses the tape analogy to illustrate the splice
operation.  The effect simulates the diagonal cuts and joins the two pieces:
.ne 14
.XE
     length1   excess
   -----------><--->
   _________   :   :  _________________
            \\  :   : :\\     `
             \\ :   : : \\     `
              \\:   : :  \\     `
               *   : :   * - - *
                \\  : :   :\\     `
                 \\ : :   : \\     `
   _______________\\: :   :  \\_____`____
                     :   :   :     :
                     <--->   <----->
                     excess  leeway
.XX
where * indicates the joining points.
.SP
For example, a long song begins with two verses which start (as
determined e.g. by using the
.B play_ng
command with the
.B trim
(\fIstart\fR) effect) at times 0:30\*d125 and 1:03\*d432.
The following commands cut out the first verse:
.XE
   sox_ng too-long.wav part1.wav trim 0 30.130
.XX
(5 ms excess, after the first verse starts)
.XE
   sox_ng too-long.wav part2.wav trim 1:03.422
.XX
(5 ms excess plus 5 ms leeway, before the second verse starts)
.XE
   sox_ng part1.wav part2.wav just-right.wav splice 30.130
.XX
For another example, the SoX command
.XE
   play_ng "|sox_ng \-n \-p synth 1 sin %1" "|sox_ng \-n \-p synth 1 sin %3"
.XX
generates and plays two notes, but there is a nasty click at the
transition; the click can be removed by splicing instead of
concatenating the audio, i.e. by appending \fBsplice 1\fR to the
command. Clicks at the beginning and end of the audio can be removed by
\fIpreceding\fR the splice effect with \fBfade q .01 2 .01\fR.
.SP
Provided your arithmetic is good enough, multiple splices can be
performed with a single
.B splice
invocation.  For example, with a Bourne shell script `acpo':
.XE
   #! /bin/sh
   # Audio Copy and Paste Over
   # acpo infile copy-start copy-stop paste-over-start outfile
   # No chained time specifications allowed for the parameters
   # (i.e. such that contain +/\-).
   e=0.005                      # Using default excess
   l=$e                         # and leeway.
   sox_ng "$1" piece.wav trim $2\-$e\-$l =$3+$e
   sox_ng "$1" part1.wav trim 0 $4+$e
   sox_ng "$1" part2.wav trim $4+$3\-$2\-$e\-$l
   sox_ng part1.wav piece.wav part2.wav "$5" \e
      splice $4+$e +$3\-$2+$e+$l+$e
.XX
two splices are used to `copy and paste' audio.
.SP
It is also possible to use this effect to perform general cross-fades,
e.g. to join two songs.  In this case,
.I excess
would typically be a number of seconds, the
.B \-q
option would typically be given to select an `equal power' cross-fade and
.I leeway
should be zero (which is the default if
.B \-q
is given).  For example, if f1.wav and f2.wav are audio files
to be cross-faded, then
.XE
   sox_ng f1.wav f2.wav out.wav splice \-q $(soxi_ng \-D f1.wav),3
.XX
cross-fades the files where the point of equal loudness is 3 seconds
before the end of f1.wav, i.e. the total length of the cross-fade is
2 \(mu 3 = 6 seconds (\f(CW$(...)\fR is POSIX shell notation that is replaced
by the output of the enclosed command).
.TP
\fBstat\fR [\fB\-s \fIscale\fR] [\fB\-rms\fR] [\fB\-freq\fR] [\fB\-v\fR] [\fB\-d\fR] [\fB\-a\fR] [\fB\-h\fR]
Display time and frequency domain statistical information about the audio.
Audio is passed unmodified through the SoX processing chain.
.SP
The information is output to the `standard error' (stderr) stream and is
calculated (where
.I n
is the duration of the audio in samples,
.I c
is the number of audio channels,
.I r
is the audio sample rate and
.I x\s-2\dk\u\s0
represents the value (in the range \-1 to +1) of each successive
sample in the audio),
as follows:
.SP
.TS
center;
lI lx.
Samples read	\fIn\fR\^\(mu\^\fIc\fR
Length (seconds)	\fIn\fR\^\(di\^\fIr\fR
Scaled by	See \fB\-s\fR below.
Maximum amplitude	T{
max(\fIx\s-2\dk\u\s0\fR)
The maximum sample value in the audio; usually this will be a positive number.
T}
Minimum amplitude	T{
min(\fIx\s-2\dk\u\s0\fR)
The minimum sample value in the audio; usually this will be a negative number.
T}
Midline amplitude	\(12\^min(\fIx\s-2\dk\u\s0\fR)\^+\^\(12\^max(\fIx\s-2\dk\u\s0\fR)
Mean norm	T{
\(S1/\s-2n\s+2\^\(*S\^\^\(br\^\fIx\s-2\dk\u\s0\fR\^\(br\^
The average of the absolute value of each sample in the audio.
T}
Mean amplitude	T{
\(S1/\s-2n\s+2\^\(*S\^\fIx\s-2\dk\u\s0\fR
The average of each sample in the audio.
If this figure is non-zero, then it indicates the
presence of a DC offset which could be removed using the
\fBdcshift\fR effect.
T}
RMS amplitude	T{
\(sr(\(S1/\s-2n\s+2\^\(*S\^\fIx\s-2\dk\u\s0\fR\(S2)
The level of a DC signal that would have the same power
as the audio's average power.
T}
Maximum delta	max(\^\(br\^\fIx\s-2\dk\u\s0\fR\^\-\^\fIx\s-2\dk\-1\u\s0\fR\^\(br\^)
Minimum delta	min(\^\(br\^\fIx\s-2\dk\u\s0\fR\^\-\^\fIx\s-2\dk\-1\u\s0\fR\^\(br\^)
Mean delta	\(S1/\s-2n\-1\s+2\^\(*S\^\^\(br\^\fIx\s-2\dk\u\s0\fR\^\-\^\fIx\s-2\dk\-1\u\s0\fR\^\(br
RMS delta	\(sr(\(S1/\s-2n\-1\s+2\^\(*S\^(\fIx\s-2\dk\u\s0\fR\^\-\^\fIx\s-2\dk\-1\u\s0\fR)\(S2)
EBUR128 Momentary	T{
The maximum momentary loudness over 400ms
T}
EBUR128 Short Term	T{
The maximum short term loudness over 3 seconds
T}
EBUR128 Integrated	T{
The integrated loudness over the whole file
T}
EBUR128 True Peak	T{
The maximum of the True Peak of each channel
T}
Rough frequency	In Hz.
Volume Adjustment	T{
The parameter to the \fBvol\fR
effect which would make the audio as loud as possible without clipping.
See the discussion on \fBClipping\fR
above for reasons why it is rarely a good idea actually to do this.
T}
.TE
.SP
Note that the delta measurements are not applicable to multichannel audio
and EBU\ R\ 128 (=ITU-R\ BS.1770) measurements are in
Loudness Units referenced to Full Scale (LUFS),
.SP
The
.B \-s
option can be used to scale the input data by a given factor.
The default value of
.I scale
is 2147483647 (the maximum value of a 32-bit signed integer)
as internal effects always work with those.
A lower value means that a different sample value should be
considered as the full-scale amplitude.
.SP
The
.B \-rms
option converts all average values to `root mean square'
format.
.SP
The
.B \-freq
option outputs the input's power spectrum (a 4096-point DFT) instead of the
statistics listed above.  This should only be used with a single-channel
audio file.
.SP
The
.B \-v
option displays only the `Volume Adjustment' value.
.SP
The
.B \-d
option
displays a hex dump of the 32-bit signed PCM data
audio in SoX's internal buffer.
This is mainly used to help track down endian problems that
sometimes occur in cross-platform versions of SoX.
.SP
The
.B \-a
option
outputs the average power spectrum instead of
the power spectrum for each 4096-point DFT.
.SP
The
.B \-h
option
uses the "histogram algorithm" to calculate the integrated EBU R-128 loudness,
which requires less memory but is less accurate.
.SP
The
.B \-j
option
outputs the statistics in JSON format, e.g.:
.XE
{
  "samples_read": 22699008,
  "length": 236.448,
  "scaled_by": 2.14748e+09,
  "maximum_amplitude": 0.818604,
  "minimum_amplitude": -0.532471,
  "midline_amplitude": 0.143066,
  "mean_norm": 0.0352694,
  "mean_amplitude": 0.00180676,
  "rms_amplitude": 0.056726,
  "maximum_delta": 0.367126,
  "minimum_delta": 0,
  "mean_delta": 0.0177341,
  "rms_delta": 0.0268538,
  "rough_frequency": 3616,
  "volume_adjustment": 1.22159
}
.XX
If \fB\-rms\fR was given, \f(CW"scaled_by"\fR will be \f(CW"scaled_by_rms"\fR
and if \fB\-e\fR was given, you also get
.XE
  "ebur128_momentary": -30.3408,
  "ebur128_short_term": -35.4501,
  "ebur128_integrated": -21.3583,
.XX
Some fields may be absent if their values are incalculable (EBUR128 figures)
or would be infinite (like the RMS of silence).
.SP
As JSON uses scientific notation, it can shows the values
of very small numbers that the usual output shows as zero.
.SP
The most common use of \fBstat\fR is to measure the characteristics
of a single audio file, for which the syntax is:
.XE
   sox_ng file.wav -n stat
.XX
where \fB\-n\fR means "No audio output is required."
.TP
\fBstats\fR [\fB\-b \fIbits\fR\^|\^\fB\-x \fIbits\fR\^|\^\fB\-s \fIscale\fR] [\fB\-w \fItime\fR] [\fB\-j\fR]
Display time domain statistical information about the audio channels;
audio is passed unmodified through the SoX processing chain.
Statistics are calculated and displayed for each audio channel and,
where applicable, an overall figure is also given.
.SP
For example, for a typical well-mastered stereo music file:
.XE
                Overall     Left      Right
   DC offset   0.000803 \-0.000391  0.000803
   Min level  \-0.750977 \-0.750977 \-0.653412
   Max level   0.708801  0.708801  0.653534
   Pk lev dB      \-2.49     \-2.49     \-3.69
   RMS lev dB    \-19.41    \-19.13    \-19.71
   RMS Pk dB     \-13.82    \-13.82    \-14.38
   RMS Tr dB     \-85.25    \-85.25    \-82.66
   Crest factor       \-      6.79      6.32
   Flat factor     0.00      0.00      0.00
   Pk count           2         2         2
   Bit-depth      16/16     16/16     16/16
   Num samples    7.72M
   Length s     174.973
   Scale max   1.000000
   Window s       0.050
.XX
.IR DC\ offset ,
.IR Min\ level ,
and
.I Max\ level
are shown, by default, in the range \(+-1.
If the
.B \-b
(bits) options is given, these three measurements are scaled to a signed integer
with the given number of bits from 2 to 32.
For example, for 16 bits, the scale would be \-32768 to +32767.
The
.B \-x
option behaves the same way as
.B \-b
except that the signed integer values are displayed in hexadecimal.
The
.B \-s
option scales the three measurements by a given floating point number.
.SP
.I Pk\ lev\ dB
and
.I RMS\ lev\ dB
are the standard peak and RMS levels measured in dBFS.
.I RMS\ Pk\ dB
and
.I RMS\ Tr\ dB
are peak and trough values of the RMS level measured over a short window
(default: 50ms).
That can be changed with the \fB\-w\fP option in seconds from 0.01 to 10.
.SP
.I Crest\ factor
is the ratio of peak to RMS level (note: not in dB).
.SP
.I Flat\ factor
is a measure of the flatness (i.e. consecutive samples with the same value) of the signal at
its peak levels (i.e. either
.I Min\ level
or
.IR Max\ level ).
.SP
.I Pk\ count
is the number of occasions (not the number of samples) that the signal attained either
.IR Min\ level ,
or
.IR Max\ level .
The primary goal of the Peak Count value is to answer the question
"has this audio been clipped?", quite possibly as a result of the
frowned-upon-by-some but common practice of 'brick wall limiting'
in modern mastering. The closer the "Peak Count" is to 1,
the higher the confidence that the audio has not been clipped.
.SP
The right-hand
.I Bit-depth
figure is the standard definition of bit-depth, i.e. that all bits other than
this number of the most significant bits are always zero.
The left-hand figure is the number of bits at the least significant end
of those most significant bits that would be sufficient
to represent all sample values accurately (including the sign bit).
.SP
In mathematical terms, the right-hand figure is the ordinal,
counting from the most significant bit,
of the least significant bit that is set to one in at least one sample.
The left-hand figure is the ordinal,
counting from the least significant repeated sign bit across all samples,
of the least significant bit that is set to one in at least one sample.
.SP
Bit-depths are not intended to be properties of the signal per se
but properties of its 2's-complement PCM encoding.
.SP
The primary use case of bit-depth measurement concerns manipulation of
PCM audio by simple bit shifting, to answer questions such as:
"Is it likely that this 24-bit PCM file was created
by simply converting a 16-bit PCM file to 24-bit?" or
"Can I losslessly shift all the samples in this PCM audio file
m-bits left or n-bits right?"
.SP
For multichannel audio, an overall figure for each of the above
measurements is given and derived from the channel figures as follows:
.IR DC\ offset :
maximum magnitude;
.IR Max\ level ,
.IR Pk\ lev\ dB ,
.IR RMS\ Pk\ dB ,
.IR Bit-depth :
maximum;
.IR Min\ level ,
.IR RMS\ Tr\ dB :
minimum;
.IR RMS\ lev\ dB ,
.IR Flat\ factor ,
.IR Pk\ count :
average;
.IR Crest\ factor :
not applicable.
.SP
.I Length\ s
is the duration in seconds of the audio and, unlike \fBstat\fP,
.I Num\ samples
is equal to the sample rate multiplied by
.IR Length .
.I Scale\ max
is the scaling applied to the first three measurements;
specifically, it is the maximum value that could apply to
.IR Max\ level .
.I Window\ s
is the length of the window used for the peak and trough RMS measurements.
.SP
The \fB\-j\fR option outputs JSON with three fields:
.RS
.TP
\f(CW"channel_count"\fR
An integer.
.TP
\f(CW"overall"\fR
An object with a member for each row of the first column of the usual output,
which are all numbers except for \f(CW"bit_depth"\fR, which is an array of two numbers.
.TP
\f(CW"channels"\fR
An array of objects with the per-channel values.
.RE
To know the overall and the channels' member names,
have a look at the output.
.SP
Like \fBstat\fR, the usual way to measure the characteristics
of a single audio file is:
.XE
   sox_ng file.wav -n stats
.XX
.TP
\fBstretch [\fIfactor\fR [\fIwindow \fR[\fIfade \fR[\fIshift \fR[\fIfading\fR]]]]]
Change the audio duration but not its pitch by cross-fading between
short windows of samples.
This effect is broadly equivalent to the
.B tempo
effect with \fIfactor\fR inverted and
.I search
set to zero so, in general, its results are comparatively poor;
it is retained as it can sometimes outperform
.B tempo
for small
.IR factor s.
.SP
.I factor
determines the change in length: >1 lengthens and <1 shortens.
By default, it is 1 (no change)
.SP
.I window
is the length of the cross-fading window in milliseconds with a default of 20.
.SP
The
.I fade
option chooses the type of crossfading:
\fBlinear\fR and \fBhalf-cosine\fR give equal-gain crossfading and cannot clip;
\fBsqrt\fR and \fBquarter-cosine\fR give two kinds of equal-power crossfading.
.SP
The
.I shift
ratio can be from 0 to 1 and its default depends on the stretch factor:
1 when speeding up, 0\*d8 when slowing down.
.SP
The
.I fading
ratio, from 0 to 0\*d5, seems to be how much of each window is cross-faded
with the adjacent ones.
The default value depends on \fIfactor\fR and \fIshift\fR:
1\*d0\ \(mi\ (\fIfactor\fR\ \(mu\ \fIshift\fR) if speeding up,
1\*d0\ \(mi\ \fIshift\fR if slowing down, with a maximum of 0\*d5.
.SP
The duration of \fBstretch\fR's output is slightly longer than
the duration of the input multiplied by \fIfactor\fR as it has to empty
the delay line it uses; \fBtempo\fR is more precise.
.TP
\fBswap\fR
Swap stereo channels.  If the input is not stereo, pairs of channels are
swapped and a possible odd last channel is passed through.  E.g., for seven
channels, the output order will be 2, 1, 4, 3, 6, 5, 7.
.SP
See
.B remix
for an effect that allows arbitrary channel selection, ordering and mixing.
.nh
.na
.TP
\fBsynth\fR [\fB\-j\^\fR\^|\^\fBp \fIkey\fR] [\fB\-n\fR] \:[\fIlength\fR [\fIoffset\fR [\fIphase\fR [\fIp1\fR [\fIp2\fR [\fIp3\fR]]]]]] \:{\fItype\fR [\fIcombine\fR [\fIfixed\fR[\fB,\fIextra\fR[\fB,\fImix\fR]]]] \:[\fIfreq\fR[\fB:\fR\^|\^\fB+\fR\^|\^\fB/\fR\^|\^\fB\-\fIfreq2\fR] \:[\fIoffset\fR \:[\fIphase\fR \:[\fIp1\fR [\fIp2\fR [\fIp3\fR]]]]]]}
.ad
.hy
.SP
\fBsynth\fR generates fixed or swept frequency audio tones
with various wave shapes and wide-band noise of various colors.
Multiple synth effects can be cascaded to produce more complex
waveforms and at each stage it is possible to choose whether the generated
waveform is mixed with or modulated onto the output of the previous stage, and
the audio for each channel in a multichannel audio file can be synthesized
independently.
.SP
It generates audio at maximum volume (0dBFS), which means that there
is a high chance of clipping so,
in many cases, you will want to follow it with the \fBgain\fR
effect to prevent this from happening. (See
.B Clipping
above.)
.SP
Though this effect is used to generate audio, an input file must still
be given, the characteristics of which are used to set the
synthesized audio length, the number of channels and the sampling rate.
However, since the input file's audio is not normally needed, a `null
file' (with the special input filename \fB\-n\fR) is often given instead
and the length specified as a parameter to \fBsynth\fR or by some other
effect that has an associated length.
.SP
By default, the tuning used
with note notations is equal temperament; the
.B \-j
.I key
option selects just intonation, where
.I key
is a whole number of semitones relative to A (so for example, \-9
or 3 selects the key of C) or a note in scientific notation and
.B \-p
selects Pythagorean tuning.
.SP
By default, the
.B synth
effect incorporates the functionality of \fBgain \-h\fR (see the
.B gain
effect for details);
.BR synth 's
.B \-n
option may be given to disable this behavior.
.SP
\fIlength\fR is the length of audio to synthesize.
A value of 0 indicated to use the input length, which is also the default.
Note that, if the input is \fB\-n\fR and the \fIlength\fR is 0 or absent,
it continues generating audio until it is stopped in some other way.
.SP
\fItype\fR is one of
.RS
.TP
.B sine
A sinusoidal wave is the default type and ignores all the \fIp\fR parameters.
.TP
.B square
A square wave.
\fIp1\fR sets the percentage of each cycle that is `on' with a default of 50.
.XE
.ne 7
  |_______        | +1
  |       |       |
  |_______|_______|  0
  |       |       |
  |       |_______| -1
  |               |
  0       p1      1
.XX
.TP
.B triangle
\fIp1\fR sets the percentage of each cycle that is `rising' with a default of 50.
.XE
.ne 7
  |    .    | +1
  |   / \e   |
  |__/___\e__|  0
  | /     \e |
  |/       \e| -1
  |         |
  0    p1   1
.XX
.TP
.B sawtooth
A sawtooth wave. With a \fIphase\fR of 0 it starts at -1 and rises to 1,
and of 10 it starts at -0.9.
The offset makes no difference.
.XE
.ne 7
  |    /| +1
  |   / |
  |__/__|  0
  | /   |
  |/    | -1
  0     1
.XX
.TP
.B trapezium
The trapezoidal wave starts at -1, rises linearly to 1, stays there,
falls linearly to -1, stays there and repeats.
\fIp1\fR sets the percentage of the cycle in which the wave is rising
with a default of 10,
\fIp2\fR sets the percentage through each cycle at which falling begins
with a default of 50 and
\fIp3\fR sets the percentage through each cycle at which falling ends
with a default of 60.
.XE
.ne 7
  |    ______             |+1
  |   /      \e            |
  |__/________\e___________| 0
  | /          \e          |
  |/            \e_________|-1
  |                       |
  0   p1    p2   p3       1
.XX
.TP
.B exp
The exponential wave rises from -1 to 1 where it peaks and immediately
begins an exponential fall.
\fIp1\fR sets the position of the maximum with a default of 50.
\fIp2\fR sets the minimum amplitude in multiples of 2dB down from the maximum
with a default of 50 (100dB);
values below 50 raise the shoulders of the wave and values above 50
lower the shoulders, increasing the pointedness of the spike.
.XE
.ne 7
  |                           | +1
  |            /\e             |
  |          _'  `_           | 0
  |        _-      -_         |
  |____---'          `---____ | f(p2)
  |                           |
  0             p1             1
.XX
.TP
.B whitenoise
Random noise with equal power at every frequency.
All noise generators ignore the \fIfrequency\fR and \fIphase\fR parameters
but if a DC offset is given, the signal's amplitude is automatically adjusted
to prevent clipping so, for noise in the range 0 to 1, an offset of 0.5 would
give a signal ranging from 0\*d0 to 1\*d0 and -0.9 from -1.0 to -0.8
.SP
\fBnoise\fR is a handy alias for \fBwhitenoise\fR
.TP
.B tpdfnoise
Noise with a Triangular Probability Density Function.
.TP
.B pinknoise
Random noise with the power at each frequency inversely proportional to the frequency.
.TP
.B brownnoise
Random noise with the power at each frequency inversely proportional to the frequency squared.
.TP
.B pluck
A plucked string simulation in which an array of sample values
representing a taut string
is set in motion with a burst of noise and decayed over time.
.SP
A plucked note's \fIfrequency\fR can be from 27.5 to 4220Hz
and the sampling rate must be between 44100 and 48000Hz.
.SP
If a DC \fIoffset\fR is used, the amplitude is automatically adjusted
to prevent clipping.
.SP
\fIp1\fR affects the sustain with a default of 40 (2dB per second);
higher values give a slower decay and lower values a faster one.
.SP
\fIp2\fR and \fIp3\fR are tone controls for the initial excitation,
with default values of 20 and 90
and a special case when \fIp3\fR is exactly 100.
If the \fIphase\fR is non-zero, it uses a different kind of random numbers.
.RE
.TP
\ 
If the \fIoffset\fR, \fIphase\fR and \fIp\fR parameters are given
before the first \fItype\fR,
they set the default values for all the following stages.
.SP
\fIcombine\fR is one of
.RS
.TP
.B create
Puts each stage's output in a new output channel and is the default:
.TP
.B mix
Mixes the generated audio 50:50 with the input signal.
.TP
.B amod
Amplitude-modulates (multiplies) the input signal by the synthesized one
considered as a value from 0 (for the most negative value)
to 1 (for the most positive value).
.TP
.B fmod
Multiplies the input signal with the synthesized one (ring modulation).
.TP
.B vdelay
Mixes the input signal with a delayed version of it
using the synthesized signal to modulate the depth of the delay.
The following three-part option \fIfixed\fR[,\fIextra\fR[,\fImix\fR]] specifies
the fixed and additional parts of the delay in milliseconds
and what percentage of the output consists of the delayed signal
from 0 for all input signal to 100 for all delayed signal
with a default of 50 (half and half).
.SP
The synthesized signal's value from \-1 to +1 varies the delay
from \fIfixed\fR seconds to \fIfixed\fR\ +\ (0\ to\ \fIextra\fR) seconds.
.SP
It interpolates linearly between the input samples and can be used
to make precision phaser, flanger and chorus-like effects, vibrato
and frequency modulation (FM) synthesis (actually phase modulation,
as used in the Yamaha DX7).
.SP
.ne 2
A chorus-like effect:
.XE
   sox_ng solo.au -d synth sine vdelay 50,2,50 .25 0 75
.XX
.ne 2
A flanger:
.XE
   sox_ng solo.au -d synth triangle vdelay 0,2,41.52 0.5 0 0
.XX
.RE
.TP
\ 
\fIfreq\fR and \fIfreq2\fR are the frequencies at the beginning and end
of the synthesis and the default frequency is 440Hz.
.SP
If
.I freq2
is given,
.I length
must also have been given and the generated tone is swept between
the given frequencies.  The two given frequencies must be separated by
one of the characters `:', `+', `/' and `\-', which
specify the sweep function as follows:
.RS
.IP \fB:\fR
Linear: the tone changes by a fixed number of hertz per second.
.IP \fB+\fR
Square: a second-order function is used to change the tone.
.IP \fB/\fR
Exponential: the tone changes by a fixed number of semitones per second.
.IP \fB\-\fR
Exponential: as `/', but the initial phase is always zero, and with stepped
(less smooth) frequency changes.
.RE
.TP
\ 
The frequency or frequency range is not used for the noise types.
.SP
\fIoffset\fR is the bias (DC offset) of the signal in percent; default=0.
.SP
\fIphase\fR is the phase shift as a percentage of 1 cycle with a default of 0
(not used for noise).
.SP
For example, the following produces a 3-second 48kHz
audio file containing a sine wave swept from 300 to 3300Hz:
.XE
   sox_ng \-n output.wav synth 3 sine 300\-3300
.XX
Multiple channels can be synthesized by specifying the set of
parameters shown between curly braces multiple times;
the following puts the swept tone in the left channel and brown
noise in the right:
.XE
   sox_ng \-n output.wav synth 3 sine 300\-3300 brownnoise
.XX
The following example shows how two synth effects can be cascaded
to create a more complex waveform:
.XE
.ne 2
   play_ng \-n synth 0.5 sine 200\-500 synth 0.5 sine fmod 700\-100
.XX
The following could be used to help tune a guitar:
.XE
.ne 2
   for n in E2 A2 D3 G3 B3 E4; do
	play_ng \-n synth 4 pluck $n repeat 2; done
.XX
.nh
.\" Otherwise it may hyphenate overlap
.TP
\fBtempo \fR[\fB\-q\fR] [\fB\-m\fR\^|\^\fBs\fR\^|\^\fBl\fR] \fIfactor\fR [\fIsegment\fR(82) [\fIsearch\fR(14.68) [\fIoverlap\fR(12)]]]
.hy
Change the audio playback speed but not its pitch. This effect uses the
WSOLA (Waveform Similarity OverLap and Add) algorithm.
The audio is chopped up into segments which are then
shifted in the time domain and overlapped (cross-faded) at points where
their waveforms are most similar as determined by the measurement of `least
squares'.
.SP
By default, linear searches are used to find the best overlapping
points. If the optional
.B \-q
parameter is given, tree searches are used instead. This makes the effect
work more quickly, but the result may not sound as good. However, if you
must improve the processing speed, this generally reduces the sound quality
less than reducing the \fIsearch\fR or \fIoverlap\fR values.
.SP
The
.B \-m
option is used to optimize the default values of \fIsegment\fR, \fIsearch\fR and
\fIoverlap\fR for music processing.
.SP
The
.B \-s
option is used to optimize default values of \fIsegment\fR, \fIsearch\fR and
\fIoverlap\fR for speech processing.
.SP
The
.B \-l
option is used to optimize default values of \fIsegment\fR, \fIsearch\fR and
\fIoverlap\fR for `linear' processing that tends to cause more
noticeable distortion but may be useful when \fIfactor\fR is close to 1.
.SP
If \fB\-m\fR, \fB\-s\fR or \fB\-l\fR is specified,
the default value of \fIsegment\fR is based on \fIfactor\fR,
while default \fIsearch\fR and \fIoverlap\fR values are based on \fIsegment\fR.
Any values you provide override these default values.
.SP
.I factor
gives the ratio of new tempo to the old tempo, so 1.1 speeds the
tempo up by 10% and 0.9 slows it down by 10%.
.SP
The optional
.I segment
parameter selects the algorithm's segment size in milliseconds.  If no other
flags are specified, the default value is 82, which is suited to
small changes in the tempo of music. For larger changes (e.g. a factor
of 2), 41 may give a better result.
The \fB\-m\fR, \fB\-s\fR, and \fB\-l\fR flags cause \fIsegment\fR's
default value to be adjusted automatically based on \fIfactor\fR.
.SP
The optional
.I search
parameter gives the audio length in milliseconds over which
the algorithm searches for overlapping points.  If no other
flags are specified, the default value is 14.68.  Larger values use
more processing time and may or may not produce better results.
A practical maximum is half the value of \fIsegment\fR. Search
can be reduced to cut processing time at the risk of degrading output
quality. The \fB\-m\fR, \fB\-s\fR and \fB\-l\fR flags cause
the search default to be adjusted automatically based on \fIsegment\fR.
.SP
The optional
.I overlap
parameter gives the segment overlap length in milliseconds.
Its default value is 12 but the \fB\-m\fR, \fB\-s\fR and \fB\-l\fR flags
automatically adjust it based on the segment size.
Increasing \fIoverlap\fR increases processing time but may increase quality.
A practical maximum for \fIoverlap\fR is a little less then \fIsearch\fR.
.SP
Note that lowering the tempo increases the sample rate and this can
make following effects slower, in particular \fBtempo\fR or \fBpitch\fR
themselves, whose running times are proportional to the sample rate times
the overlap, all squared.
This can be compensated for by following \fBtempo\fR with
a fast \fBrate\fR effect.
.SP
See \fBspeed\fR for an effect that changes tempo and pitch together,
\fBpitch\fR and \fBbend\fR for effects that change pitch only and
\fBstretch\fR for an effect that changes the tempo using a different algorithm.
.TP
\fBtreble \fIgain\fR [\fIfrequency\fR [\fIwidth\fR[\fBs\fR\^|\^\fBh\fR\^|\^\fBk\fR\^|\^\fBo\fR\^|\^\fBq\fR]]]
Apply a treble tone control effect.
See the description of the \fBbass\fR effect for details.
.TP
\fBtremolo \fIspeed\fR [\fIdepth\fR]
Apply a tremolo (low frequency sinusoidal amplitude modulation)
effect to the audio.
The frequency of the tremolo in Hz is given by \fIspeed\fR
and its \fIdepth\fR is a percentage with a default of 40.
.TP
\fBtrim\fR {\fIposition(+)\fR}
Cuts out portions of the audio.  Any number of \fIposition\fRs may be
given; audio is not sent to the output until the first \fIposition\fR
is reached.  The effect then alternates between copying and discarding
audio at each \fIposition\fR.  Using a value of 0 for the first \fIposition\fR
parameter allows copying from the beginning of the audio.
.SP
For example,
.XE
   sox_ng in.au out.au trim 0 10
.XX
copies the first ten seconds, while
.XE
   play_ng in.au trim 12:34 =15:00 -2:00
.XX
and
.XE
   play_ng in.au trim 12:34 2:26 -2:00
.XX
both play from 12 minutes 34 seconds into the audio up to 15 minutes in
(i.e. 2 minutes and 26 seconds long) then resume playing two
minutes before the end.
.SP
SoX has an internal speed hack which, when \fBtrim\fR is the first effect
and removes audio from the beginning, seeks in the audio file instead of
decoding it and throwing the data away but this is only used when
the input is a single file. To achieve fast gapless playing with
multiple files and trimming the first, you can use something like:
.XE
   FMT='-t s32 -r 44100 -c2'
   (sox file1.mp3 $FMT - trim 30 && sox file2.mp3 $FMT -) | play $FMT -
.XX
.TP
\fBupsample\fR [\fIfactor(2)\fR]
Upsample the signal by an integer factor: \fIfactor\fR\-1 zero-valued
samples are inserted between each pair of input samples.  As a result,
the original spectrum is replicated into the new frequency space and
attenuated.  This attenuation can be compensated for by adding
\fBvol\ \fIfactor\fR.
The \fBupsample\fR effect is typically used in combination with filtering effects.
.SP
For a general resampling effect with antialiasing, see \fBrate\fR.
See \fBdownsample\fR.
.TP
\fBvad \fR[\fIoptions\fR]
The Voice Activity Detector attempts to trim silence and quiet
background sounds from the ends of (fairly high resolution
i.e. 16-bit, 44\-48kHz) recordings of speech.  The algorithm currently
uses a simple cepstral power measurement to detect voice, so may be
fooled by other things, especially music.  The effect can trim only
from the front of the audio, so in order to trim from the back, the
.B reverse
effect must also be used.  E.g.
.XE
   play_ng speech.wav norm vad
.XX
to trim from the front,
.XE
   play_ng speech.wav norm reverse vad reverse
.XX
to trim from the back and
.XE
   play_ng speech.wav norm vad reverse vad reverse
.XX
to trim from both ends.  The use of the
.B norm
effect is recommended, but remember that neither
.B reverse
nor
.B norm
is suitable for use with streamed audio.
.TP
\ 
.B Options
.br
Default values are shown in parentheses, the allowed range in square brackets.
.RS
.IP "\fB\-t\ \fInum\fR (7) [0 \- 20]"
The measurement level used to trigger activity detection.  This might
need to be changed depending on the noise level, signal level and
other characteristics of the input audio.
.IP "\fB\-T \fInum\fR (0.25) [0.01 \- 1]"
The time constant (in seconds) used to help ignore short bursts of
sound.
.IP "\fB\-s \fInum\fR (1) [0.1 \- 4]"
The amount of audio (in seconds) to search for quieter/shorter bursts
of audio to include prior to the detected trigger point.
.IP "\fB\-g \fInum\fR (0.25) [0.1 \- 1]"
Allowed gap (in seconds) between quieter/shorter bursts of audio to
include prior to the detected trigger point.
.IP "\fB\-p \fInum\fR (0) [0 \- 4]"
The amount of audio (in seconds) to preserve before the trigger point
and any found quieter/shorter bursts.
.RE
.SP
There are keymaps on \fItrigger_level\fR, \fItrigger_time\fR and \fIgap\fR.
.TP
\ 
.B Advanced Options
.br
These allow fine tuning of the algorithm's internal parameters.
.RS
.IP "\fB\-b \fInum\fR (0.35) [0.1 \- 10]"
The algorithm uses adaptive noise estimation/reduction in
order to detect the start of the wanted audio.
This option sets the time in seconds for the initial noise estimate.
.IP "\fB\-N \fInum\fR (0.1) [0.1 \- 10]"
Time constant used by the adaptive noise estimator when the noise
level is increasing.
.IP "\fB\-n \fInum\fR (0.01) [0.001 \- 0.1]"
Time constant used by the adaptive noise estimator when the noise
level is decreasing.
.IP "\fB\-r \fInum\fR (1.35) [0 \- 2]"
Amount of noise reduction to use in the detection algorithm.
.IP "\fB\-f \fInum\fR (20) [5 \- 50]"
Frequency of the algorithm's processing/measurements.
.IP "\fB\-m \fInum\fR (0.1) [0.01 \- 1]"
Measurement duration. By default, it is twice the measurement period;
i.e. with 50% overlap, but if you set \fB-f\fR,
you also need to change \fB\-m\fR to 2 divided by its value
to keep a 50% overlap.
.IP "\fB\-M \fInum\fR (0.4) [0-1 \- 1]"
Time constant used to smooth spectral measurements.
.IP "\fB\-h \fIfreq\fR (50) [10 \-\^]"
`Brick-wall' frequency of the high-pass filter applied at the
detector algorithm's input.
.IP "\fB\-l \fIfreq\fR (6000) [1000 \-\^]"
`Brick-wall' frequency of the low-pass filter applied at the
detector algorithm's input.
.IP "\fB\-H \fIfreq\fR (150) [10 \-\^]"
`Brick-wall' frequency of the high-pass lifter used in the detector algorithm.
.IP "\fB\-L \fIfreq\fR (2000) [1000 \-\^]"
`Brick-wall' frequency of the low-pass lifter used in the detector algorithm.
.RE
.TP
\ 
See the
.B silence
effect.
.TP
\fBvol \fIgain\fR [\fItype\fR [\fIlimiter\-gain\fR]]
Apply amplification or attenuation to the audio signal.
Unlike
.BR \-v ,
which is used for balancing multiple input files as they enter the
SoX effects processing chain,
.B vol
is an effect like any other so can be applied anywhere in the processing chain
and several times if necessary.
.SP
The amount to change the volume is given by
.I gain
which is interpreted, according to the given \fItype\fR, as follows: if
.I type
is \fBamplitude\fR (or is omitted),
.I gain
is an amplitude ratio (voltage or linear),
if \fBpower\fR, a power ratio (wattage or voltage squared)
and if \fBdB\fR, a power change in dB.
.SP
When
.I type
is \fBamplitude\fR or \fBpower\fR, a
.I gain
of 1 leaves the volume unchanged,
less than 1 decreases it,
and greater than 1 increases it;
a negative
.I gain
inverts the audio signal in addition to adjusting its volume.
.SP
When
.I type
is \fBdB\fR, a
.I gain
of 0 leaves the volume unchanged,
less than 0 decreases it
and greater than 0 increases it.
.SP
See [4]
for a detailed discussion on electrical (and hence audio signal)
voltage and power ratios.
.SP
Beware of
.B Clipping
when the increasing the volume.
.SP
The
.I gain
and the
.I type
parameters can be concatenated if desired, e.g.
.BR "vol 10dB" .
.SP
An optional \fIlimiter\-gain\fR value can be specified and should be a
value much less
than 1 (e.g. 0\*d05 or 0\*d02) and is used only on peaks to prevent clipping.
Not specifying this parameter causes no limiter to be used.  In verbose
mode, this effect displays the percentage of the audio that needed to be
limited.
.SP
There is a keymap on \fIvol.gain\fR, which is adjusted in the
units that were specified (amplitude, dB or power).
.SP
See
.B gain
for a volume-changing effect with different capabilities and
.B compand
for a dynamic range compression/expansion/limiting effect.
.SH REFERENCES
.TP
[1]
R. Bristow-Johnson,
.IR "Cookbook formulae for audio EQ biquad filter coefficients" ,
.br
https://www.w3.org/TR/audio-eq-cookbook
.TP
[2]
Wikipedia,
.IR "Q-factor" ,
.br
http://en.wikipedia.org/wiki/Q_factor
.TP
[3]
Scott Lehman,
.IR "Effects Explained" ,
.br
https://codeberg.org/sox_ng/Effects-Explained
.TP
[4]
Wikipedia,
.IR "Decibel" ,
.br
http://en.wikipedia.org/wiki/Decibel
.TP
[5]
Richard Furse,
.IR "Linux Audio Developer's Simple Plugin API" ,
.br
http://www.ladspa.org
.TP
[6]
Richard Furse,
.IR "Computer Music Toolkit" ,
.br
http://www.ladspa.org/cmt/overview.html
.TP
[7]
Steve Harris,
.IR "LADSPA plugins" ,
.br
http://plugin.org.uk
.SH SEE ALSO
.BR sox_ng (1).
.SH AUTHORS
Lance Norskog, Chris Bagwell and many other authors and contributors
listed in the README file that is distributed with the source code.