• audioop —- Manipulate raw audio data

    audioop —- Manipulate raw audio data

    The audioop module contains some useful operations on sound fragments.It operates on sound fragments consisting of signed integer samples 8, 16, 24or 32 bits wide, stored in bytes-like objects. All scalar items areintegers, unless specified otherwise.

    在 3.4 版更改: Support for 24-bit samples was added.All functions now accept any bytes-like object.String input now results in an immediate error.

    This module provides support for a-LAW, u-LAW and Intel/DVI ADPCM encodings.

    A few of the more complicated operations only take 16-bit samples, otherwise thesample size (in bytes) is always a parameter of the operation.

    The module defines the following variables and functions:

    • exception audioop.error

    • This exception is raised on all errors, such as unknown number of bytes persample, etc.

    • audioop.add(fragment1, fragment2, width)

    • Return a fragment which is the addition of the two samples passed as parameters.width is the sample width in bytes, either 1, 2, 3 or 4. Bothfragments should have the same length. Samples are truncated in case of overflow.

    • audioop.adpcm2lin(adpcmfragment, width, state)

    • Decode an Intel/DVI ADPCM coded fragment to a linear fragment. See thedescription of lin2adpcm() for details on ADPCM coding. Return a tuple(sample, newstate) where the sample has the width specified in width.

    • audioop.alaw2lin(fragment, width)

    • Convert sound fragments in a-LAW encoding to linearly encoded sound fragments.a-LAW encoding always uses 8 bits samples, so width refers only to the samplewidth of the output fragment here.

    • audioop.avg(fragment, width)

    • Return the average over all samples in the fragment.

    • audioop.avgpp(fragment, width)

    • Return the average peak-peak value over all samples in the fragment. Nofiltering is done, so the usefulness of this routine is questionable.

    • audioop.bias(fragment, width, bias)

    • Return a fragment that is the original fragment with a bias added to eachsample. Samples wrap around in case of overflow.

    • audioop.byteswap(fragment, width)

    • "Byteswap" all samples in a fragment and returns the modified fragment.Converts big-endian samples to little-endian and vice versa.

    3.4 新版功能.

    • audioop.cross(fragment, width)

    • Return the number of zero crossings in the fragment passed as an argument.

    • audioop.findfactor(fragment, reference)

    • Return a factor F such that rms(add(fragment, mul(reference, -F))) isminimal, i.e., return the factor with which you should multiply reference tomake it match as well as possible to fragment. The fragments should bothcontain 2-byte samples.

    The time taken by this routine is proportional to len(fragment).

    • audioop.findfit(fragment, reference)

    • Try to match reference as well as possible to a portion of fragment (whichshould be the longer fragment). This is (conceptually) done by taking slicesout of fragment, using findfactor() to compute the best match, andminimizing the result. The fragments should both contain 2-byte samples.Return a tuple (offset, factor) where offset is the (integer) offset intofragment where the optimal match started and factor is the (floating-point)factor as per findfactor().

    • audioop.findmax(fragment, length)

    • Search fragment for a slice of length length samples (not bytes!) withmaximum energy, i.e., return i for which rms(fragment[i2:(i+length)2])is maximal. The fragments should both contain 2-byte samples.

    The routine takes time proportional to len(fragment).

    • audioop.getsample(fragment, width, index)

    • Return the value of sample index from the fragment.

    • audioop.lin2adpcm(fragment, width, state)

    • Convert samples to 4 bit Intel/DVI ADPCM encoding. ADPCM coding is an adaptivecoding scheme, whereby each 4 bit number is the difference between one sampleand the next, divided by a (varying) step. The Intel/DVI ADPCM algorithm hasbeen selected for use by the IMA, so it may well become a standard.

    state is a tuple containing the state of the coder. The coder returns a tuple(adpcmfrag, newstate), and the newstate should be passed to the next callof lin2adpcm(). In the initial call, None can be passed as the state.adpcmfrag is the ADPCM coded fragment packed 2 4-bit values per byte.

    • audioop.lin2alaw(fragment, width)

    • Convert samples in the audio fragment to a-LAW encoding and return this as abytes object. a-LAW is an audio encoding format whereby you get a dynamicrange of about 13 bits using only 8 bit samples. It is used by the Sun audiohardware, among others.

    • audioop.lin2lin(fragment, width, newwidth)

    • Convert samples between 1-, 2-, 3- and 4-byte formats.

    注解

    In some audio formats, such as .WAV files, 16, 24 and 32 bit samples aresigned, but 8 bit samples are unsigned. So when converting to 8 bit widesamples for these formats, you need to also add 128 to the result:

    1. new_frames = audioop.lin2lin(frames, old_width, 1)
    2. new_frames = audioop.bias(new_frames, 1, 128)

    The same, in reverse, has to be applied when converting from 8 to 16, 24or 32 bit width samples.

    • audioop.lin2ulaw(fragment, width)

    • Convert samples in the audio fragment to u-LAW encoding and return this as abytes object. u-LAW is an audio encoding format whereby you get a dynamicrange of about 14 bits using only 8 bit samples. It is used by the Sun audiohardware, among others.

    • audioop.max(fragment, width)

    • Return the maximum of the absolute value of all samples in a fragment.

    • audioop.maxpp(fragment, width)

    • Return the maximum peak-peak value in the sound fragment.

    • audioop.minmax(fragment, width)

    • Return a tuple consisting of the minimum and maximum values of all samples inthe sound fragment.

    • audioop.mul(fragment, width, factor)

    • Return a fragment that has all samples in the original fragment multiplied bythe floating-point value factor. Samples are truncated in case of overflow.

    • audioop.ratecv(fragment, width, nchannels, inrate, outrate, state[, weightA[, weightB]])

    • Convert the frame rate of the input fragment.

    state is a tuple containing the state of the converter. The converter returnsa tuple (newfragment, newstate), and newstate should be passed to the nextcall of ratecv(). The initial call should pass None as the state.

    The weightA and weightB arguments are parameters for a simple digital filterand default to 1 and 0 respectively.

    • audioop.reverse(fragment, width)

    • Reverse the samples in a fragment and returns the modified fragment.

    • audioop.rms(fragment, width)

    • Return the root-mean-square of the fragment, i.e. sqrt(sum(S_i^2)/n).

    This is a measure of the power in an audio signal.

    • audioop.tomono(fragment, width, lfactor, rfactor)

    • Convert a stereo fragment to a mono fragment. The left channel is multiplied bylfactor and the right channel by rfactor before adding the two channels togive a mono signal.

    • audioop.tostereo(fragment, width, lfactor, rfactor)

    • Generate a stereo fragment from a mono fragment. Each pair of samples in thestereo fragment are computed from the mono sample, whereby left channel samplesare multiplied by lfactor and right channel samples by rfactor.

    • audioop.ulaw2lin(fragment, width)

    • Convert sound fragments in u-LAW encoding to linearly encoded sound fragments.u-LAW encoding always uses 8 bits samples, so width refers only to the samplewidth of the output fragment here.

    Note that operations such as mul() or max() make no distinctionbetween mono and stereo fragments, i.e. all samples are treated equal. If thisis a problem the stereo fragment should be split into two mono fragments firstand recombined later. Here is an example of how to do that:

    1. def mul_stereo(sample, width, lfactor, rfactor):
    2.     lsample = audioop.tomono(sample, width, 1, 0)
    3.     rsample = audioop.tomono(sample, width, 0, 1)
    4.     lsample = audioop.mul(lsample, width, lfactor)
    5.     rsample = audioop.mul(rsample, width, rfactor)
    6.     lsample = audioop.tostereo(lsample, width, 1, 0)
    7.     rsample = audioop.tostereo(rsample, width, 0, 1)
    8.     return audioop.add(lsample, rsample, width)

    If you use the ADPCM coder to build network packets and you want your protocolto be stateless (i.e. to be able to tolerate packet loss) you should not onlytransmit the data but also the state. Note that you should send the _initial_state (the one you passed to lin2adpcm()) along to the decoder, not thefinal state (as returned by the coder). If you want to usestruct.Struct to store the state in binary you can code the firstelement (the predicted value) in 16 bits and the second (the delta index) in 8.

    The ADPCM coders have never been tried against other ADPCM coders, only againstthemselves. It could well be that I misinterpreted the standards in which casethey will not be interoperable with the respective standards.

    The find*() routines might look a bit funny at first sight. They areprimarily meant to do echo cancellation. A reasonably fast way to do this is topick the most energetic piece of the output sample, locate that in the inputsample and subtract the whole output sample from the input sample:

    1. def echocancel(outputdata, inputdata):
    2.     pos = audioop.findmax(outputdata, 800)    # one tenth second
    3.     out_test = outputdata[pos*2:]
    4.     in_test = inputdata[pos*2:]
    5.     ipos, factor = audioop.findfit(in_test, out_test)
    6.     # Optional (for better cancellation):
    7.     # factor = audioop.findfactor(in_test[ipos*2:ipos*2+len(out_test)],
    8.     #              out_test)
    9.     prefill = '\0'*(pos+ipos)*2
    10.     postfill = '\0'*(len(inputdata)-len(prefill)-len(outputdata))
    11.     outputdata = prefill + audioop.mul(outputdata, 2, -factor) + postfill
    12.     return audioop.add(inputdata, outputdata, 2)