Faster-Whisper 本地实时语音转文本部署指南

Faster-Whisper 本地实时语音转文本部署指南

要实现类似主流应用的语音输入功能，通常有两种方案：云端 API（轻量、准确）和本地模型（免费、隐私、离线）。本文记录使用 Faster-Whisper 实现本地实时语音识别的完整流程。

环境准备

在虚拟环境中安装依赖。注意：标准库为 pyaudio，原教程中的 pyaudiowpatch 应为笔误，此处已修正。

pip install faster-whisper pyaudio torch

若需 GPU 加速，请确保系统已正确安装 CUDA 和 cuDNN。

模型下载

支持多种模型规格，根据资源情况选择：

• Tiny / Base / Small: 速度快，资源占用低
• Medium / Large-v2 / Large-v3: 精度高，适合对准确率要求高的场景
• Distil-Large-v3: 蒸馏版，兼顾速度与效果

模型文件可从 Hugging Face 获取。手动下载时，请将以下关键文件放入同一目录：

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

实时录音转文本脚本

以下是核心代码示例。脚本实现了音频录制、切片处理及模型转录，并针对 GPU 推理进行了优化。

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

# 录音切片时长（秒）
AUDIO_BUFFER = 5

def record_audio(p, device):
    """创建临时 WAV 文件进行录音"""
    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f:
        filename = f.name
        wave_file = wave.open(filename, "wb")
        wave_file.setnchannels(int(device[]))
        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()