#!/usr/bin/env python3 """ KB Whisper - 本地语音识别工具 基于 Whisper.cpp 的音视频转文字 """ import argparse import json import os import subprocess import sys from pathlib import Path import re def parse_arguments(): """解析命令行参数""" parser = argparse.ArgumentParser( description='KB Whisper - 本地语音识别工具', formatter_class=argparse.RawDescriptionHelpFormatter, epilog=""" 示例: python whisper.py --input video.mp4 python whisper.py --input video.mp4 --model base python whisper.py --input audio.mp3 --output ./output --background """ ) parser.add_argument( '--input', required=True, help='输入音视频文件路径' ) parser.add_argument( '--model', default='tiny', choices=['tiny', 'base', 'small', 'medium', 'large'], help='Whisper 模型类型 (默认: tiny)' ) parser.add_argument( '--output', help='输出目录 (默认: 输入文件所在目录)' ) parser.add_argument( '--background', action='store_true', help='后台异步执行' ) parser.add_argument( '--language', help='指定语言 (默认: 自动检测),例如: en, zh, ja' ) return parser.parse_args() def get_whisper_cpp_path(): """获取 whisper.cpp 路径""" # 优先从环境变量获取 path = os.environ.get('WHISPER_CPP_PATH') if path: path = os.path.expanduser(path) if os.path.exists(path): return path # 尝试常见位置 common_paths = [ os.path.expanduser('~/whisper.cpp'), '/usr/local/bin/whisper', '/opt/whisper.cpp', ] for cp in common_paths: if os.path.exists(cp): return cp raise FileNotFoundError( "未找到 whisper.cpp。请设置环境变量 WHISPER_CPP_PATH " "或安装到 ~/whisper.cpp\n" "安装方法: git clone https://github.com/ggerganov/whisper.cpp.git ~/whisper.cpp && cd ~/whisper.cpp && make" ) def check_model(whisper_path, model): """检查模型是否存在""" model_path = os.path.join(whisper_path, 'models', f'ggml-{model}.bin') if not os.path.exists(model_path): raise FileNotFoundError( f"模型文件不存在: {model_path}\n" f"请运行以下命令下载模型:\n" f"cd {whisper_path} && ./models/download-ggml-model.sh {model}" ) return model_path def run_whisper_cpp(whisper_path, input_file, model, output_dir, language=None): """运行 whisper.cpp 命令行工具""" # 构建命令 cmd = [ os.path.join(whisper_path, 'main'), '-m', os.path.join(whisper_path, 'models', f'ggml-{model}.bin'), '-f', input_file, '-of', os.path.join(output_dir, Path(input_file).stem), '-otxt', # 输出 txt '-osrt', # 输出 srt '-oj', # 输出 json ] # 添加语言参数(如果指定) if language: cmd.extend(['-l', language]) print(f"执行命令: {' '.join(cmd)}") # 运行命令 result = subprocess.run( cmd, capture_output=True, text=True, cwd=whisper_path ) if result.returncode != 0: raise RuntimeError( f"Whisper.cpp 执行失败:\n" f"标准输出: {result.stdout}\n" f"错误输出: {result.stderr}" ) return result def parse_whisper_json(json_file): """解析 whisper.cpp 输出的 JSON 文件""" try: with open(json_file, 'r', encoding='utf-8') as f: data = json.load(f) return data except json.JSONDecodeError: # whisper.cpp 的 JSON 格式可能不同,尝试解析标准格式 return None def create_standard_json(input_file, output_dir, srt_file, txt_file, model): """创建标准格式的 JSON 输出""" basename = Path(input_file).stem # 解析 SRT 文件获取时间戳 segments = [] if os.path.exists(srt_file): with open(srt_file, 'r', encoding='utf-8') as f: content = f.read() # 解析 SRT 格式 pattern = r'(\d+)\n(\d{2}:\d{2}:\d{2},\d{3}) --> (\d{2}:\d{2}:\d{2},\d{3})\n(.+?)(?=\n\n|\Z)' matches = re.findall(pattern, content, re.DOTALL) for idx, (num, start, end, text) in enumerate(matches): # 转换时间戳 def parse_time(t): h, m, s_ms = t.split(':') s, ms = s_ms.split(',') return int(h) * 3600 + int(m) * 60 + int(s) + int(ms) / 1000 segments.append({ "id": idx, "start": parse_time(start), "end": parse_time(end), "text": text.strip() }) # 读取纯文本 txt_content = "" if os.path.exists(txt_file): with open(txt_file, 'r', encoding='utf-8') as f: txt_content = f.read().strip() # 获取音频时长(使用 ffprobe) duration = 0 try: result = subprocess.run( ['ffprobe', '-v', 'error', '-show_entries', 'format=duration', '-of', 'default=noprint_wrappers=1:nokey=1', input_file], capture_output=True, text=True ) if result.returncode == 0: duration = float(result.stdout.strip()) except FileNotFoundError: pass # 构建标准 JSON json_data = { "success": True, "input_file": os.path.basename(input_file), "input_path": input_file, "model": model, "output_files": { "json": f"{basename}.json", "srt": f"{basename}.srt", "txt": f"{basename}.txt" }, "duration_seconds": duration, "segments": segments } return json_data def main(): """主函数""" args = parse_arguments() # 检查输入文件 input_file = os.path.abspath(args.input) if not os.path.exists(input_file): print(f"错误: 输入文件不存在: {input_file}", file=sys.stderr) sys.exit(1) # 获取输出目录 if args.output: output_dir = os.path.abspath(args.output) else: output_dir = os.path.dirname(input_file) os.makedirs(output_dir, exist_ok=True) # 获取 whisper.cpp 路径 try: whisper_path = get_whisper_cpp_path() print(f"Whisper.cpp 路径: {whisper_path}") except FileNotFoundError as e: print(f"错误: {e}", file=sys.stderr) sys.exit(1) # 检查模型 try: check_model(whisper_path, args.model) print(f"使用模型: {args.model}") except FileNotFoundError as e: print(f"错误: {e}", file=sys.stderr) sys.exit(1) # 后台执行 if args.background: print("启动后台处理...") # 使用 nohup 在后台运行 log_file = os.path.join(output_dir, f"{Path(input_file).stem}_whisper.log") with open(log_file, 'w') as log: subprocess.Popen( [sys.executable, __file__, '--input', input_file, '--model', args.model, '--output', output_dir] + (['--language', args.language] if args.language else []), stdout=log, stderr=log, start_new_session=True ) print(f"后台任务已启动,日志文件: {log_file}") print(f"输出文件将生成到: {output_dir}") # 返回任务信息 result = { "success": True, "background": True, "status": "running", "log_file": log_file, "output_dir": output_dir, "message": "后台处理中,请稍候检查输出文件" } print(json.dumps(result, ensure_ascii=False, indent=2)) return # 前台执行 print(f"处理文件: {input_file}") print(f"输出目录: {output_dir}") try: # 运行 whisper.cpp result = run_whisper_cpp( whisper_path, input_file, args.model, output_dir, args.language ) # 输出文件路径 basename = Path(input_file).stem json_file = os.path.join(output_dir, f"{basename}.json") srt_file = os.path.join(output_dir, f"{basename}.srt") txt_file = os.path.join(output_dir, f"{basename}.txt") # 检查输出文件 output_files = {} if os.path.exists(srt_file): output_files['srt'] = srt_file if os.path.exists(txt_file): output_files['txt'] = txt_file # 创建标准 JSON 格式 standard_json = create_standard_json( input_file, output_dir, srt_file, txt_file, args.model ) # 写入标准 JSON with open(json_file, 'w', encoding='utf-8') as f: json.dump(standard_json, f, ensure_ascii=False, indent=2) output_files['json'] = json_file # 输出结果 result = { "success": True, "input_file": input_file, "model": args.model, "output_files": { k: os.path.basename(v) for k, v in output_files.items() }, "output_dir": output_dir, "total_segments": len(standard_json.get('segments', [])), "duration": standard_json.get('duration_seconds', 0) } print("\n✓ 处理完成!") print(f"\n输出文件:") for fmt, fname in output_files.items(): print(f" - {fname.upper()}: {fname}") print(f"\n完整结果:") print(json.dumps(result, ensure_ascii=False, indent=2)) except Exception as e: print(f"\n✗ 错误: {e}", file=sys.stderr) sys.exit(1) if __name__ == '__main__': main()