Search code examples
pythonaudiobase64wavconverters

how to pass edited wav between functions without saving wav in between?


I have a wav conversation of 2 people(customer and tech support) I have 3 separate functions that extract 1 voice, cut 10 seconds and transform it to embedding.

def get_customer_voice(file):

    print('getting customer voice only')
    wav = wf.read(file)
    ch = wav[1].shape[1]#customer voice always in 1st track
    sr = wav[0]
    c1 = wav[1][:,1]
    #print('c0 %i'%c0.size)

    if ch==1:
        exit()
    vad = VoiceActivityDetection()
    vad.process(c1)
    voice_samples = vad.get_voice_samples()
    #this is trouble - how to pass it without saving anywhere as wav?
    wf.write('%s_customer.wav'%file,sr,voice_samples)

function below cuts 10 seconds of wav file from function above.

import sys
from pydub import AudioSegment

def get_customer_voice_10_seconds(file):
    voice = AudioSegment.from_wav(file)
    new_voice = voice[0:10000]
    file = str(file) + '_10seconds.wav'
    new_voice.export(file, format='wav')


if __name__ == '__main__':
    if len(sys.argv) < 2:
        print('give wav file to process!')
    else:
        print(sys.argv)
        get_customer_voice_10_seconds(sys.argv[1])

how to pass it as wav or other format without saving it to some directory? It's to be used in rest api, i don't know where it will save that wav, so preferably it should be passed somehow.


Solution

  • I figured it out - the function below just works without saving, buffer etc. It receives a wav file and edits it and just sends straight to the get math embedding function:

    def get_customer_voice_and_cutting_10_seconds_embedding(file):
    
        print('getting customer voice only')
        wav = read(file)
        ch = wav[1].shape[1]
        sr = wav[0]
    
        c1 = wav[1][:,1]
    
        vad = VoiceActivityDetection()
        vad.process(c1)
        voice_samples = vad.get_voice_samples()
        audio_segment = AudioSegment(voice_samples.tobytes(), frame_rate=sr,sample_width=voice_samples.dtype.itemsize, channels=1)
        audio_segment = audio_segment[0:10000]
        file = str(file) + '_10seconds.wav'
    
        return get_embedding(file)
    

    the key is tobytes() in Audio segment, it just assembles all them together in 1 track again