audio.py 文件源码

def extract(self, filename, **kwargs):
        speech = ''

        # convert to wav, if not already .wav
        base, ext = os.path.splitext(filename)
        if ext != '.wav':
            temp_filename = self.convert_to_wav(filename)
            try:
                speech = self.extract(temp_filename, **kwargs)
            finally:  # make sure temp_file is deleted
                os.remove(temp_filename)
        else:
            r = sr.Recognizer()

            with sr.WavFile(filename) as source:
                audio = r.record(source)

            try:
                speech = r.recognize_google(audio)
            except LookupError:  # audio is not understandable
                speech = ''

            # add a newline, to make output cleaner
            speech += '\n'

        return speech