利用OpenAI Whisper将音频转换为SRT字幕,支持Intel XPU、NVIDIA CUDA、AMD ROCm及Apple Metal的自动GPU加速,适用于内容创作者。
Convert audio files to SRT subtitles using local Whisper models — completely free, offline, and GPU accelerated.
DeviceAccelerationFP16Intel Arc SeriesXPU❌ Auto disabledNVIDIA GPUsCUDA✅ Auto enabledAMD GPUsROCm✅ Auto enabledApple M SeriesMetal✅ Auto enabledNo GPUCPU❌ Auto disabled
Place the audio file in your current working directory and tell the AI:
text
Convert xxx.mp3 to SRT subtitles
Or specify the full path directly:
text
Convert /path/to/audio.mp3 to SRT subtitles
text
Convert xxx.mp3 to English subtitles using large-v3-turbo model
Convert xxx.mp3 to subtitles, language is Japanese
AI will execute the scripts/transcribe.py script, which will:
turbo)pip install torch==2.10.0+xpupip install torch --index-url https://download.pytorch.org/whl/cu121pip install torchpip install openai-whisper~/.cache/whisper by default, use symlink/Junction to redirect to another diskTip for China users: If model download fails, manually download from mirror sites and place in
~/.cache/whisper/
ModelSizeSpeedAccuracytiny39MFastestLowbase74MFastMediumsmall244MMediumMediummedium769MSlowHighturbo809MMediumHigh ✅ Recommendedlarge-v31550MSlowestHighestlarge-v3-turbo1550MSlowHighest
使用本地 Whisper 模型将音频文件转录为 SRT 字幕, 完全免费,无需联网,支持 GPU 加速。
设备加速方式FP16Intel Arc 系列XPU❌ 自动禁用NVIDIA 显卡CUDA✅ 自动启用AMD 显卡ROCm✅ 自动启用Apple M 系列Metal✅ 自动启用无独显CPU❌ 自动禁用
将音频文件放入当前工作目录,然后告诉 AI:
text
把 xxx.mp3 转成 SRT 字幕文件
或者直接指定路径:
text
把 /path/to/audio.mp3 转成 SRT 字幕
text
把 xxx.mp3 用 large-v3-turbo 模型转成英文字幕
把 xxx.mp3 转成字幕,语言是日语
AI 会调用 scripts/transcribe.py 脚本执行转录,脚本会:
turbo)pip install torch==2.10.0+xpupip install torch --index-url https://download.pytorch.org/whl/cu121pip install torchpip install openai-whisper 自动安装~/.cache/whisper,可用软链接/Junction 指向其他磁盘国内用户提示:首次运行会自动下载模型,如下载失败可手动从镜像站下载后放入
~/.cache/whisper/
模型大小速度准确度tiny39M最快低base74M快中small244M中中medium769M慢高turbo809M中高 ✅ 推荐large-v31550M最慢最高large-v3-turbo1550M慢最高