Faster-Whisper 本地部署实时语音转文本教程

前言

要实现语音输入功能，通常有两种主流方案：云端 API（轻量、准确度高）和本地模型（免费、隐私、无需联网）。本文记录使用 Faster-Whisper 实现本地实时语音输入转文本的部署过程。

一、安装环境

在虚拟环境中安装 faster-whisper 和 pyaudio：

pip install faster-whisper
pip install pyaudio

若需使用 GPU 加速，请确保已正确安装 CUDA 和 cuDNN。

二、使用步骤

1. 下载模型

如需离线使用或指定模型路径，可手动从 Hugging Face 下载。根据需求选择模型版本：

• Tiny (最小/最快)
• Base
• Small
• Medium
• Large-v2
• Large-v3 (效果最好)
• Distil-Large-v3 (蒸馏版/速度快)

在 Hugging Face 的 Files and versions 页面中，下载以下关键文件并放入同一文件夹：

• config.json
• model.bin
• tokenizer.json
• vocabulary.json
• preprocessor_config.json

2. 实时录音转文本脚本

# -*- coding: utf-8 -*-
import os
import sys
import time
import wave
import tempfile
import threading
import torch
import pyaudio
from faster_whisper import WhisperModel

AUDIO_BUFFER = 5

def record_audio(p, device):
    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f:
        filename = f.name
        wave_file = wave.open(filename, "wb")
        wave_file.setnchannels(int(device["maxInputChannels"]))
        wave_file.setsampwidth(p.get_sample_size(pyaudio.paInt16))
        wave_file.setframerate((device[]))

         ():
            wave_file.writeframes(in_data)
             (in_data, pyaudio.paContinue)

        :
            stream = p.(=pyaudio.paInt16,
                            channels=(device[]),
                            rate=(device[]),
                            frames_per_buffer=,
                            =,
                            input_device_index=device[],
                            stream_callback=callback,)
            time.sleep(AUDIO_BUFFER)
         Exception  e:
            ()
        :
               ():
                stream.stop_stream()
                stream.close()
            wave_file.close()
         filename

 ():
    :
        segments, info = model.transcribe(
            filename,
            beam_size=,
            language=,
            vad_filter=,
            vad_parameters=(min_silence_duration_ms=)
        )
         segment  segments:
            ( % (segment.start, segment.end, segment.text))
     Exception  e:
        ()
    :
         os.path.exists(filename):
            os.remove(filename)

 ():
    ()
     torch.cuda.is_available():
        device = 
        compute_type = 
        ()
    :
        device = 
        compute_type = 
        ()

    model_path = 
    :
        model = WhisperModel(model_path, device=device, compute_type=compute_type, local_files_only=)
        ()
     Exception  e:
        ()
        

     pyaudio.PyAudio()  p:
        :
            default_mic = p.get_default_input_device_info()
            ()
            ()
            ( * )
            ()
             :
                filename = record_audio(p, default_mic)
                thread = threading.Thread(target=whisper_audio, args=(filename, model))
                thread.start()
         OSError:
            ()
         KeyboardInterrupt:
            ()
         Exception  e:
            ()

 __name__ == :
    main()