Creating suitable WAV files for Google Speech API

Question

I'm using pyaudio to record my voice as wav file. I'm using following code:

def voice_recorder():
    FORMAT = pyaudio.paInt16
    CHANNELS = 2
    RATE = 22050
    CHUNK = 1024
    RECORD_SECONDS = 4
    WAVE_OUTPUT_FILENAME = "first.wav"

    audio = pyaudio.PyAudio()

    # start Recording
    stream = audio.open(format=FORMAT, channels=CHANNELS,
                    rate=RATE, input=True,
                    frames_per_buffer=CHUNK)
    print "konusun..."
    frames = []

    for i in range(0, int(RATE / CHUNK * RECORD_SECONDS)):
        data = stream.read(CHUNK)
        frames.append(data)
    #print "finished recording"


    # stop Recording
    stream.stop_stream()
    stream.close()
    audio.terminate()

    waveFile = wave.open(WAVE_OUTPUT_FILENAME, 'wb')
    waveFile.setnchannels(CHANNELS)
    waveFile.setsampwidth(audio.get_sample_size(FORMAT))
    waveFile.setframerate(RATE)
    waveFile.writeframes(b''.join(frames))
    waveFile.close()

I'm using following code for Google Speech API which basically converts the speech in the WAV file to text: https://github.com/GoogleCloudPlatform/python-docs-samples/blob/master/speech/api-client/transcribe.py

When I try to import the wav file which is generated by pyaudio to Google's code, I'm getting following error:

googleapiclient.errors.HttpError: <HttpError 400 when requesting https://speech.googleapis.com/v1beta1/speech:syncrecognize?alt=json returned "Invalid Configuration, Does not match Wav File Header.
Wav Header Contents:
Encoding: LINEAR16
Channels: 2
Sample Rate: 22050.
Request Contents:
Encoding: linear16
Channels: 1
Sample Rate: 22050.">

I'm using following workaround for this: I'm converting WAV file to MP3 with ffmpeg, after then I'm converting MP3 file to wav again with sox:

def wav_to_mp3():
    FNULL = open(os.devnull, 'w')
    subprocess.call(['ffmpeg', '-i', 'first.wav', '-ac', '1', '-ab', '6400', '-ar', '16000', 'second.mp3', '-y'], stdout=FNULL, stderr=subprocess.STDOUT)

def mp3_to_wav():
    subprocess.call(['sox', 'second.mp3', '-r', '16000', 'son.wav'])

Google's API works with this WAV output but since quality decreases too much, it doesn't perform well.

So how can I create Google compatible WAV file with pyaudio at the first step?

Answer 1

Converting wav file to flac file with avconv and sending it to Google Speech API solved the problem

subprocess.call(['avconv', '-i', 'first.wav', '-y', '-ar', '48000', '-ac', '1', 'last.flac'])