record.py 文件源码

def listen():
    with speech_recognition.Microphone() as source:
        recognizer.adjust_for_ambient_noise(source)
        audio = recognizer.listen(source)

    try:
        return recognizer.recognize_sphinx(audio)
        # or: return recognizer.recognize_google(audio)
    except speech_recognition.UnknownValueError:
        print("Could not understand audio")
    except speech_recognition.RequestError as e:
        print("Recog Error; {0}".format(e))

    return ""