for i in range(0, 8):
file = filename + "channel" + str(i) + ".wav"
WAV_FILE = path.join(path.dirname(path.realpath(__file__)), file)
with sr.WavFile(WAV_FILE) as source:
audio = r.record(source) # read the entire WAV file
# recognize speech using Google Speech Recognition
try:
# for testing purposes, we're just using the default API key
# to use another API key, use `r.recognize_google(audio, key="GOOGLE_SPEECH_RECOGNITION_API_KEY")`
# instead of `r.recognize_google(audio)`
testtext = r.recognize_google(audio)
print("Google Speech Recognition for mic " + str(i) + "::::::::::" + str(testtext.encode('utf-8')))
res = wer.wer(reftext, testtext)
print('Word Error Rate: {0:.04f}'.format(res))
except sr.UnknownValueError:
print("Google Speech Recognition could not understand audio")
except sr.RequestError as e:
print("Could not request results from Google Speech Recognition service; {0}".format(e))
WAV_FILE = path.join(path.dirname(path.realpath(__file__)), filename + "SD.wav")
with sr.WavFile(WAV_FILE) as source:
audio = r.record(source) # read the entire WAV file
# recognize speech using Google Speech Recognition
try:
# for testing purposes, we're just using the default API key
# to use another API key, use `r.recognize_google(audio, key="GOOGLE_SPEECH_RECOGNITION_API_KEY")`
f1 = '/home/tyf/api/api_test/testdata/测试文书/传唤证/常玮平传唤通知.jpg_2018-08-15_15-37-23.txt' f2 = '/home/tyf/api/api_test/testdata/测试文书/传唤证/恥松.txt' # f1='E:\项目\\18.7.16东大项目\测试文书\' # f1='E:\项目\\18.7.16东大项目\测试文书\拘留证\陈鸿志-黑社会.txt' # f2='E:\项目\\18.7.16东大项目\测试文书\拘留证\据留.txt' fref = codecs.open(f1, 'r', encoding='utf-8').readlines() fpred = codecs.open(f2, 'r', encoding='utf-8').readlines() fref = [s.strip() for s in fref] fpred = [s.strip() for s in fpred] fref = ''.join(fref) fpred = ''.join(fpred) # print(fref) # print(fref[0]) ret = asr_wer.wer(fref, fpred, debug=True) # print lcs(fref,fpred) # fref= u' '.join([char for char in fref]) # fpred= u' '.join([char for char in fpred]) # # ret = asr_wer.wer(fref,fpred, debug=True) #fref=[u' '.join(list(s.strip().replace(u' ',u''))) for s in fref ] #fpred=[u' '.join(list(s.strip().replace(u' ',u''))) for s in fpred ] # err_numSub, err_numDel , err_numIns,total_char=0,0,0,0 # # sentcount=0 # # for r,p in zip(fref,fpred): # numSub, numDel , numIns,tchars=wer.wer(r,p) # err_numSub+=numSub