pyaudio+pocketsphinx

参考例程：https://pythonism.wordpress.com/2013/06/06/pocketsphinx-voice-recognition-with-python/

          
其他参考：http://blog.youkuaiyun.com/u013000753/article/details/46344567
https://github.com/mondhs/lt-pocketsphinx-tutorial/tree/master/impl/demo-py/public_service 立陶宛Lithuanian
功能点：pyaudio录音。音频流。
importsys,os
importpyaudio
importwave

 

hmdir= "/usr/share/pocketsphinx/model/hmm/wsj1"

lmd  = "/usr/share/pocketsphinx/model/lm/wsj/wlist5o.3e-7.vp.tg.lm.DMP"

dictd= "/usr/share/pocketsphinx/model/lm/wsj/wlist5o.dic"

 

defdecodeSpeech(hmmd,lmdir,dictp,wavfile):

 

    importpocketsphinx as ps

    importsphinxbase

 

    speechRec= ps.Decoder(hmm = hmmd, lm = lmdir, dict = dictp)

    wavFile= file(wavfile,'rb')

    wavFile.seek(44)

    speechRec.decode_raw(wavFile)

    result= speechRec.get_hyp()

 

    returnresult[0]

 

CHUNK= 1024

FORMAT= pyaudio.paInt16

CHANNELS= 1

RATE= 16000

RECORD_SECONDS= 10

 

forx inrange(10):

    fn= "o"+str(x)+".wav"

    p= pyaudio.PyAudio()

    stream= p.open(format=FORMAT, channels=CHANNELS, rate=RATE,input=True, frames_per_buffer=CHUNK)

    print("*recording")

    frames= []

    fori inrange(0,int(RATE/ CHUNK * RECORD_SECONDS)):

        data= stream.read(CHUNK)

        frames.append(data)

    print("*done recording")

    stream.stop_stream()

    stream.close()

    p.terminate()

    wf= wave.open(fn,'wb')

    wf.setnchannels(CHANNELS)

    wf.setsampwidth(p.get_sample_size(FORMAT))

    wf.setframerate(RATE)

    wf.writeframes(b''.join(frames))

    wf.close()

    wavfile= fn

    recognised= decodeSpeech(hmdir,lmd,dictd,wavfile)

    printrecognised

    cm= 'espeak "'+recognised+'"'

    os.system(cm)