Question

我是Python的新手，他试图弄清楚如何实时从文件中转录音频语音，并在后台播放声音。

更新：

@petezurich对不起，很抱歉。目前，我可以听到音频在后台播放。但是，我很难 Sphinx抄录音频。方式有问题吗我正在将音频传递给Sphinx？它一直在输出“ Sphinx错误”消息。

我正在将 PocketSpinx 与 Uberi / speech_recognition 库一起使用。

到目前为止，这是我整理的内容：

#!/usr/bin/env python
# recognitions.py : Transcribe Test from an Audio File

import os
import sys
import time
import wave
import pyaudio
import speech_recognition as sr
import threading

try:
    import pocketsphinx
except:
    print("PocketSphinx is not installed.")

# import audio file within script folder
from os import path
audio_file = path.join(os.path.abspath(os.path.dirname(sys.argv[0])), "samples/OSR_us_000_0061_8k.wav")
print("Transcribing... " + audio_file)
wf = wave.open(audio_file, 'rb')

# set PyAudio instance
pa = pyaudio.PyAudio()
# set recognizer instance (unmodified)
r = sr.Recognizer()

stream_buffer = bytes()
stream_counter = 0
audio_sampling_rate = 48000

def main_recognize(stream):
    global audio_sampling_rate
    # Create a new AudioData instance, which represents "mono" audio data
    audio_data = sr.AudioData(stream, audio_sampling_rate, 2)

    # recognize using CMU Sphinx (en-US only)
    try:
        print("Sphinx: " + r.recognize_sphinx(audio_data, language="en-US"))
    except sr.UnknownValueError:
        print("Sphinx error")
    except sr.RequestError as e:
        print("Sphinx error; {0}".format(e))

def stream_audio(data):
    global stream_buffer
    global stream_counter

    buffer_set_size = 200

    if stream_counter < buffer_set_size:
        # force 'data' to BYTES to allow concat
        data = bytes()
        stream_buffer += data
        stream_counter += 1
    else:
        threading.Thread(target=main_recognize, args=(stream_buffer,)).start()
        # reset
        stream_buffer = bytes()
        stream_counter = 0


# define callback
def callback(in_data, frame_count, time_info, status):
    data = wf.readframes(frame_count)
    stream_audio(in_data)
    return (data, pyaudio.paContinue)
# open audio stream
stream = pa.open(format=pa.get_format_from_width(wf.getsampwidth()),
                channels=wf.getnchannels(),
                rate=wf.getframerate(),
                output=True,
                stream_callback=callback)
# start the stream
stream.start_stream()
# wait for stream to finish
while stream.is_active():
    time.sleep(0.1)
# stop stream
stream.stop_stream()
stream.close()
wf.close()
# close PyAudio
pa.terminate()

关于我可能做错了什么的建议或建议？

我的方法是否朝着正确的方向前进？

提前谢谢！

https://github.com/Uberi/speech_recognition/blob/master/reference/library-reference.rst

Answer 1

Uberi包装器不适用于流，您应该尝试使用original pocketsphinx API之类的东西

config = Decoder.default_config()
config.set_string('-hmm', path.join(MODELDIR, 'en-us/en-us'))
config.set_string('-lm', path.join(MODELDIR, 'en-us/en-us.lm.bin'))
config.set_string('-dict', path.join(MODELDIR, 'en-us/cmudict-en-us.dict'))
config.set_string('-logfn', '/dev/null')
decoder = Decoder(config)

stream = open(path.join(DATADIR, 'goforward.raw'), 'rb')
#stream = open('10001-90210-01803.wav', 'rb')

in_speech_bf = False
decoder.start_utt()
while True:
    buf = stream.read(1024)
    if buf:
        decoder.process_raw(buf, False, False)
        if decoder.get_in_speech() != in_speech_bf:
            in_speech_bf = decoder.get_in_speech()
            if not in_speech_bf:
                decoder.end_utt()
                print 'Result:', decoder.hyp().hypstr
                decoder.start_utt()
    else:
        break
decoder.end_utt()

使用python speech_recognition播放和流式传输音频

1 个答案: