Python 3.5.2

Ran a few small tests today using py-webrtcvad with a 19 second WAV file. I'm impressed with the results as the output WAV's appear to respect what I will call 'word boundaries'. That is, words are not broken up on output.

Can an enhancement be made to limit the audio file chunk sizes, as the audio chunk files created were between 1 second and 48 seconds. Apparently smaller WAV files are required for training purposes in Speech Recognition.

Hi, I was curious whether we could use this tool for real time application. That is whether we could detect voices coming directly from a mic in a noisy environment. If so I would be glad if anyone could give me or direct me to such implementation.

Dear Wiseman, When I use webrtcvad2.0.10 to detect some wav,I found the result is different when I comment the code

    num_voiced = len([f for f in ring_buffer
                              if vad.is_speech(f.bytes, sample_rate)])

env: centos6.5 Python 2.7.14

before comment:

000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000

after comment:

000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001111100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000

If you get your test files with whatever audio I'm generating, everything works. Each chunk of his was in the "size" of 480 (variable n). If you put any other value outside of this, the application receives the following error:

chunk = file.read(960)
>>> vad.is_speech(chunk, 8000)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python3.6/dist-packages/webrtcvad.py", line 27, in is_speech
    return _webrtcvad.process(self._vad, sample_rate, buf, length)
webrtcvad.Error: Error while processing frame

• Synced webrtc with the latest master at 05aee741f696a96c6f7b931d9a84e92aa51ecbde from upstream.
• Added python versions 3.6 to 3.7 to travis.
• Fixed test_valid_rate_and_frame_length by catching OverflowError as well (This failed for me on win10 python 3.7 before I made any modifications): OverflowError: Python int too large to convert to C long

I confirmed that all tests passed in python 3.6.5 (Linux) and python 3.7.0 (Windows).

Im working on an IBM Power VM and there seems to be no wheel for it. How would I build one myself if there is none?

Python 3.6 Ubuntu 18

Thanks for a great project

   ERROR: Command errored out with exit status 1:
     command: /home/anaconda3/envs/wmlce-1.6.1/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-m8vmxfbt/webrtcvad/setup.py'"'"'; __file__='"'"'/tmp/pip-install-m8vmxfbt/webrtcvad/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' install --record /tmp/pip-record-zhq51ldg/install-record.txt --single-version-externally-managed --compile
         cwd: /tmp/pip-install-m8vmxfbt/webrtcvad/
    Complete output (24 lines):
    running install
    running build
    running build_py
    creating build
    creating build/lib.linux-ppc64le-3.6
    copying webrtcvad.py -> build/lib.linux-ppc64le-3.6
    running build_ext
    building '_webrtcvad' extension
    creating build/temp.linux-ppc64le-3.6
    creating build/temp.linux-ppc64le-3.6/cbits
    creating build/temp.linux-ppc64le-3.6/cbits/webrtc
    creating build/temp.linux-ppc64le-3.6/cbits/webrtc/common_audio
    creating build/temp.linux-ppc64le-3.6/cbits/webrtc/common_audio/signal_processing
    creating build/temp.linux-ppc64le-3.6/cbits/webrtc/common_audio/vad
    gcc -pthread -B /home/anaconda3/envs/wmlce-1.6.1/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -DWEBRTC_POSIX -Icbits -I/home/anaconda3/envs/wmlce-1.6.1/include/python3.6m -c cbits/pywebrtcvad.c -o build/temp.linux-ppc64le-3.6/cbits/pywebrtcvad.o
    In file included from cbits/webrtc/common_audio/vad/include/webrtc_vad.h:19:0,
                     from cbits/pywebrtcvad.c:2:
    cbits/webrtc/typedefs.h:51:2: error: #error Please add support for your architecture in typedefs.h
     #error Please add support for your architecture in typedefs.h
      ^~~~~
    cbits/webrtc/typedefs.h:55:2: error: #error Define either WEBRTC_ARCH_LITTLE_ENDIAN or WEBRTC_ARCH_BIG_ENDIAN
     #error Define either WEBRTC_ARCH_LITTLE_ENDIAN or WEBRTC_ARCH_BIG_ENDIAN
      ^~~~~
    error: command 'gcc' failed with exit status 1
    ----------------------------------------
ERROR: Command errored out with exit status 1: /home/anaconda3/envs/wmlce-1.6.1/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-m8vmxfbt/webrtcvad/setup.py'"'"'; __file__='"'"'/tmp/pip-install-m8vmxfbt/webrtcvad/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' install --record /tmp/pip-record-zhq51ldg/install-record.txt --single-version-externally-managed --compile Check the logs for full command output.

Hi, I had been trying to read the various .wav files with different python libraries. I tried using wave as you did and also tried librosa, ffmpeg and soundfile. But with different libraries i am getting different number of chunks in the output. Can you please explain why is it that VAD works the best when we use wave reader over the other audio readers.

I am running example.py as it is on my Windows 10 machine with Python 3.6. The sample wav file being used is extracted from a video recorded on a phone. There is no speech frames in this 8 seconds audio but still the output from the example.py is exactly like the input and seem to mark most of the frames as is_speech = 1. I am puzzled as the output should have not marked any frame as speech! Could there be some issue with this wav file?

Sample wav file 11.wav

Executed with params - example.py 3 11.wav

Terminal output is as below: 00001111111111+(0.12)1111111111111111111111111110111111111111111111111011100111101111111111111111111011110000000011111111111111111111111111111111111111111111111111101111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111-(8.280000000000015) Writing chunk-00.wav

I have an 8-bit recording but it seems like the webrtcvad only supports 16-bit recordings. Is there any way to get 8-bit recordings to work other than converting them to 16-bit representation?

I am using vad algorithm by modifying the 'example.py' code you have uploaded. While running the code, I found something strange.

I added some lines to the example.py to print is_speech() result of the same frame for several times.

def vad_collector(sample_rate, frame_duration_ms, padding_duration_ms, vad, frames):
    num_padding_frames = int(padding_duration_ms / frame_duration_ms)
    ring_buffer = collections.deque(maxlen=num_padding_frames)
    triggered = False
    voiced_frames = []
    for frame in frames:
        sys.stdout.write(
            '1' if vad.is_speech(frame.bytes, sample_rate) else '0')
        if not triggered:
            ring_buffer.append(frame)


            # print vad results
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            print(vad.is_speech(ring_buffer[0].bytes, sample_rate))
            import os;os.exit()        # stop the code


            num_voiced = len([f for f in ring_buffer
                              if vad.is_speech(f.bytes, sample_rate)])
            if num_voiced > 0.9 * ring_buffer.maxlen:
                sys.stdout.write('+(%s)' % (ring_buffer[0].timestamp,))
                triggered = True
                voiced_frames.extend(ring_buffer)
                ring_buffer.clear()
        else:
            voiced_frames.append(frame)
            ring_buffer.append(frame)
            num_unvoiced = len([f for f in ring_buffer
                                if not vad.is_speech(f.bytes, sample_rate)])
            if num_unvoiced > 0.9 * ring_buffer.maxlen:
                sys.stdout.write('-(%s)' % (frame.timestamp + frame.duration))
                triggered = False
                yield b''.join([f.bytes for f in voiced_frames])
                ring_buffer.clear()
                voiced_frames = []
    if triggered:
        sys.stdout.write('-(%s)' % (frame.timestamp + frame.duration))
    sys.stdout.write('\n')
    if voiced_frames:
        yield b''.join([f.bytes for f in voiced_frames])

I thought every printed line should show the same boolean value (True or False), but there were different values among them. Could you explain this result?

I couldn't install it by pip on windows, and received the error messages below;

error: command 'C:\Users\XXX\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\cl.exe' failed with exit status 2

If somebody suggest me, it would be great helpful! Hiro

I'm having the hardest time figuring out how to convert standard PyAudio frames into your VAD format. I feel like there should be an example for something this basic. I saw your post about using PyAudio Mic in, and how you have confirmed it works. #29

I'm trying to adapt the frame_generator, but am coming up short. My logic is to use PyAudio to determine if the input is "Silent" or "Not Silent". After a noise is detected and the logic drops back into "Silent" mode, then we run the cached audio frames through VAD.

Something like this...

         FORMAT = pyaudio.paInt16
         CHANNELS = 1  # 2
         RATE = 48000
         NUM_SAMPLES = 1024
   
           self.stream = self.p.open(
            format=self.settings.FORMAT,
            channels=self.settings.CHANNELS,
            rate=self.settings.RATE,
            # frames_per_buffer=
            frames_per_buffer=self.settings.NUM_SAMPLES,
            input=True,
            input_device_index=self.settings.INPUT_DEVICE_INDEX
        )

The RMS detect logic

           if mean > silence and in_silence:
                print("Sound!")
                in_silence = False
                
            if mean > silence and not in_silence:
                frames.append(frame)  # the list to eventually send to VAD

            elif mean < silence and not in_silence:
                print("Silence!")
                in_silence = True
                self.doVadDetect(frames, self.settings.RATE)

doVadDetect function is pretty much copy pasta - what am I missing in the conversion??

 def doVadDetect(self, audio_frames, sample_rate):
        frames = self.frame_generator(10, audio_frames, sample_rate)
        frames = list(frames)
        print("frames: " + str(len(frames)))
        segments = self.vad_collector(sample_rate, 30, 300, self.vad, frames)
        for i, segment in enumerate(segments):
            path = 'chunk-%002d.wav' % (i,)
            print(' Writing %s' % (path,))
            self.write_wave(path, segment, sample_rate)

Thanks for the help!

Sir, Madam,

I may need some help in order to use webrtcvad in multiprocess environment. I'm working with pcm alaw audio which are converted in linear pcm for analysis purpose. Before processing the analysis, I'm using webrtcvad library in order to detect silences. Everything is ok if I'm doing one analysis at a time but as long I'm doing several silence detection at a time it fails.

The following code detect the silences:

import asyncio
import audioop
import concurrent.futures
import os
import webrtcvad

n = int(16000 * (30 / 1000.0) * 2)
vad = webrtcvad.Vad(2)

file1_name = "44022"
file2_name = "44024"

def process_chunk(data, file):
    while len(data) >= n:
        audio_30ms = data[0:n]
        data = data[n:]
        is_speech = vad.is_speech(audio_30ms, 16000)
        print(f"{file} - Is speech? {'true' if is_speech else 'false'}")


async def file1(pool):
    loop = asyncio.get_running_loop()
    f = open(f"{file1_name}-1.raw", "rb")
    data = f.read(960)
    while data:
        audioop.alaw2lin(data, 2)
        data = audioop.ratecv(data, 2, 1, 8000, 16000, None)[0]
        await loop.run_in_executor(pool, process_chunk, data, file1_name)
        data = f.read(960)


async def file2(pool):
    loop = asyncio.get_running_loop()
    f = open(f"{file2_name}-1.raw", "rb")
    data = f.read(960)
    while data:
        audioop.alaw2lin(data, 2)
        data = audioop.ratecv(data, 2, 1, 8000, 16000, None)[0]
        await loop.run_in_executor(pool, process_chunk, data, file2_name)
        data = f.read(960)


async def test():
    pool = concurrent.futures.ThreadPoolExecutor((os.cpu_count() or 1))
    await file1(pool)
    await file2(pool)


if __name__ == '__main__':
    asyncio.run(test())

The following doesn't detect the silences

import asyncio
import audioop
import concurrent.futures
import os
import webrtcvad

n = int(16000 * (30 / 1000.0) * 2)
vad = webrtcvad.Vad(2)

file1_name = "44022"
file2_name = "44024"

def process_chunk(data, file):
    while len(data) >= n:
        audio_30ms = data[0:n]
        data = data[n:]
        is_speech = vad.is_speech(audio_30ms, 16000)
        print(f"{file} - Is speech? {'true' if is_speech else 'false'}")


async def file1(pool):
    loop = asyncio.get_running_loop()
    f = open(f"{file1_name}-1.raw", "rb")
    data = f.read(960)
    while data:
        audioop.alaw2lin(data, 2)
        data = audioop.ratecv(data, 2, 1, 8000, 16000, None)[0]
        await loop.run_in_executor(pool, process_chunk, data, file1_name)
        data = f.read(960)


async def file2(pool):
    loop = asyncio.get_running_loop()
    f = open(f"{file2_name}-1.raw", "rb")
    data = f.read(960)
    while data:
        audioop.alaw2lin(data, 2)
        data = audioop.ratecv(data, 2, 1, 8000, 16000, None)[0]
        await loop.run_in_executor(pool, process_chunk, data, file2_name)
        data = f.read(960)


async def test():
    pool = concurrent.futures.ThreadPoolExecutor((os.cpu_count() or 1))
    await asyncio.gather(file1(pool), file2(pool))


if __name__ == '__main__':
    asyncio.run(test())

Did I miss something?

I also added some logs to display the buffer which is analized, and in both case, the buffer is the same

audio.zip

Below is the error i got. please give some advice

C:\WINDOWS\system32>pip install webrtcvad Collecting webrtcvad Using cached webrtcvad-2.0.10.tar.gz (66 kB) Preparing metadata (setup.py) ... done Using legacy 'setup.py install' for webrtcvad, since package 'wheel' is not installed. Installing collected packages: webrtcvad Running setup.py install for webrtcvad ... error error: subprocess-exited-with-error

× Running setup.py install for webrtcvad did not run successfully. │ exit code: 1 ╰─> [9 lines of output] running install running build running build_py creating build creating build\lib.win-amd64-3.9 copying webrtcvad.py -> build\lib.win-amd64-3.9 running build_ext building '_webrtcvad' extension error: Microsoft Visual C++ 14.0 or greater is required. Get it with "Microsoft C++ Build Tools": https://visualstudio.microsoft.com/visual-cpp-build-tools/ [end of output]

note: This error originates from a subprocess, and is likely not a problem with pip. error: legacy-install-failure

× Encountered error while trying to install package. ╰─> webrtcvad

note: This is an issue with the package mentioned above, not pip. hint: See above for output from the failure.

ERROR: Command errored out with exit status 1: command: 'C:\Users\DappExpert\anaconda3\envs\clone\python.exe' -u -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'C:\Users\DAPPEX~1\AppData\Local\Temp\pip-install-rnoq_ipo\webrtcvad_66a6a7f27ca44e97b589eb9ceb91fe4a\setup.py'"'"'; file='"'"'C:\Users\DAPPEX~1\AppData\Local\Temp\pip-install-rnoq_ipo\webrtcvad_66a6a7f27ca44e97b589eb9ceb91fe4a\setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(file) if os.path.exists(file) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' bdist_wheel -d 'C:\Users\DAPPEX~1\AppData\Local\Temp\pip-wheel-9__nm872' cwd: C:\Users\DAPPEX~1\AppData\Local\Temp\pip-install-rnoq_ipo\webrtcvad_66a6a7f27ca44e97b589eb9ceb91fe4a
Complete output (19 lines): running bdist_wheel running build running build_py creating build creating build\lib.win-amd64-3.6 copying webrtcvad.py -> build\lib.win-amd64-3.6 running build_ext building '_webrtcvad' extension creating build\temp.win-amd64-3.6 creating build\temp.win-amd64-3.6\Release creating build\temp.win-amd64-3.6\Release\cbits creating build\temp.win-amd64-3.6\Release\cbits\webrtc creating build\temp.win-amd64-3.6\Release\cbits\webrtc\common_audio creating build\temp.win-amd64-3.6\Release\cbits\webrtc\common_audio\signal_processing creating build\temp.win-amd64-3.6\Release\cbits\webrtc\common_audio\vad C:\Program Files (x86)\Microsoft Visual Studio\2019\BuildTools\VC\Tools\MSVC\14.29.30133\bin\HostX86\x64\cl.exe /c /nologo /Ox /W3 /GL /DNDEBUG /MD -D_WIN32 -Icbits -IC:\Users\DappExpert\anaconda3\envs\clone\include -IC:\Users\DappExpert\anaconda3\envs\clone\include "-IC:\Program Files (x86)\Microsoft Visual Studio\2019\BuildTools\VC\Tools\MSVC\14.29.30133\include" /Tccbits\pywebrtcvad.c /Fobuild\temp.win-amd64-3.6\Release\cbits\pywebrtcvad.obj pywebrtcvad.c C:\Users\DappExpert\anaconda3\envs\clone\include\pyconfig.h(59): fatal error C1083: ЌҐ г¤ Ґвбп ®вЄалвм д ©« ўЄ«о祭ЁҐ: io.h: No such file or directory, error: command 'C:\Program Files (x86)\Microsoft Visual Studio\2019\BuildTools\VC\Tools\MSVC\14.29.30133\bin\HostX86\x64\cl.exe' failed with exit status 2

ERROR: Failed building wheel for webrtcvad Running setup.py clean for webrtcvad Failed to build webrtcvad Installing collected packages: zipp, typing-extensions, urllib3, threadpoolctl, scipy, pyparsing, llvmlite, joblib, importlib-metadata, idna, dataclasses, charset-normalizer, cached-property, werkzeug, scikit-learn, requests, protobuf, packaging, numba, markdown, jsonpointer, importlib-resources, h5py, grpcio, colorama, appdirs, absl-py, wrapt, websocket-client, tqdm, tornado, torchfile, termcolor, tensorflow-estimator, tensorboard, soundfile, resampy, pyzmq, python-dateutil, PyQt5-sip, PyQt5-Qt5, pynndescent, pooch, pillow, kiwisolver, keras-preprocessing, keras-applications, jsonpatch, google-pasta, gast, dill, decorator, cycler, audioread, astor, webrtcvad, visdom, Unidecode, umap-learn, tensorflow-gpu, sounddevice, PyQt5, multiprocess, matplotlib, librosa, inflect, distance Running setup.py install for webrtcvad ... error ERROR: Command errored out with exit status 1: command: 'C:\Users\DappExpert\anaconda3\envs\clone\python.exe' -u -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'C:\Users\DAPPEX~1\AppData\Local\Temp\pip-install-rnoq_ipo\webrtcvad_66a6a7f27ca44e97b589eb9ceb91fe4a\setup.py'"'"'; file='"'"'C:\Users\DAPPEX~1\AppData\Local\Temp\pip-install-rnoq_ipo\webrtcvad_66a6a7f27ca44e97b589eb9ceb91fe4a\setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(file) if os.path.exists(file) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' install --record 'C:\Users\DAPPEX~1\AppData\Local\Temp\pip-record-p11aszvu\install-record.txt' --single-version-externally-managed --compile --install-headers 'C:\Users\DappExpert\anaconda3\envs\clone\Include\webrtcvad' cwd: C:\Users\DAPPEX~1\AppData\Local\Temp\pip-install-rnoq_ipo\webrtcvad_66a6a7f27ca44e97b589eb9ceb91fe4a
Complete output (19 lines): running install running build running build_py creating build creating build\lib.win-amd64-3.6 copying webrtcvad.py -> build\lib.win-amd64-3.6 running build_ext building '_webrtcvad' extension creating build\temp.win-amd64-3.6 creating build\temp.win-amd64-3.6\Release creating build\temp.win-amd64-3.6\Release\cbits creating build\temp.win-amd64-3.6\Release\cbits\webrtc creating build\temp.win-amd64-3.6\Release\cbits\webrtc\common_audio creating build\temp.win-amd64-3.6\Release\cbits\webrtc\common_audio\signal_processing creating build\temp.win-amd64-3.6\Release\cbits\webrtc\common_audio\vad C:\Program Files (x86)\Microsoft Visual Studio\2019\BuildTools\VC\Tools\MSVC\14.29.30133\bin\HostX86\x64\cl.exe /c /nologo /Ox /W3 /GL /DNDEBUG /MD -D_WIN32 -Icbits -IC:\Users\DappExpert\anaconda3\envs\clone\include -IC:\Users\DappExpert\anaconda3\envs\clone\include "-IC:\Program Files (x86)\Microsoft Visual Studio\2019\BuildTools\VC\Tools\MSVC\14.29.30133\include" /Tccbits\pywebrtcvad.c /Fobuild\temp.win-amd64-3.6\Release\cbits\pywebrtcvad.obj pywebrtcvad.c C:\Users\DappExpert\anaconda3\envs\clone\include\pyconfig.h(59): fatal error C1083: ЌҐ г¤ Ґвбп ®вЄалвм д ©« ўЄ«о祭ЁҐ: io.h: No such file or directory, error: command 'C:\Program Files (x86)\Microsoft Visual Studio\2019\BuildTools\VC\Tools\MSVC\14.29.30133\bin\HostX86\x64\cl.exe' failed with exit status 2 ---------------------------------------- ERROR: Command errored out with exit status 1: 'C:\Users\DappExpert\anaconda3\envs\clone\python.exe' -u -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'C:\Users\DAPPEX~1\AppData\Local\Temp\pip-install-rnoq_ipo\webrtcvad_66a6a7f27ca44e97b589eb9ceb91fe4a\setup.py'"'"'; file='"'"'C:\Users\DAPPEX~1\AppData\Local\Temp\pip-install-rnoq_ipo\webrtcvad_66a6a7f27ca44e97b589eb9ceb91fe4a\setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(file) if os.path.exists(file) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' install --record 'C:\Users\DAPPEX~1\AppData\Local\Temp\pip-record-p11aszvu\install-record.txt' --single-version-externally-managed --compile --install-headers 'C:\Users\DappExpert\anaconda3\envs\clone\Include\webrtcvad' Check the logs for full command output.

Memory leak problem still existed in version 2.0.10, but it is said thath it has been fixed in 2.0.10 according to README.md. If you run the following code, you will found that the memory of process quickly increase.

for _ in range(100000):
        webrtcvad.Vad(3)