Starstruk Logo MarkStarstruk

v3.66

jianchang512/pyvideotransv3.66Mar 18, 2025by jianchang512

AI Summary

Introduced OpenAI GPT-4o models for transcription and TTS, along with hardware acceleration optimizations.

Key Highlights

  • Added support for `gpt-4o-mini-tts` and `gpt-4o-transcribe`.
  • Optimized hardware acceleration for video processing.
  • Upgraded F5-TTS to use the official interface.

New Features

  • GPT-4o model support
  • Hardware acceleration optimization

Full Release Notes

> 预打包版仅适用于Windows10/11，MacOS和Linux请源码部署
> MacOS/Linux升级：重新拉取源码覆盖，然后执行
>  `pip3 install --upgrade openai-whisper elevenlabs` 
> `pip install --upgrade gradio_client`
> `pip3 install --no-deps  --force-reinstall "faster-whisper @ https://github.com/SYSTRAN/faster-whisper/archive/refs/heads/master.tar.gz"`

## Change

- Feat: support openai gpt-4o-mini-tts / gpt-4o-transcribe
- Fix: 优化视频处理时硬件加速
- Fix: 升级F5-TTS接口 [使用官方接口](https://pvt9.com/f5tts)


## Win v3.67 完整包下载

> 如果未安装过旧版本，请在此下载完整版，如需cuda加速，需有英伟达显卡并且安装cuda12.x及cudnn9

百度网盘下载地址(含tiny/medium模型):   https://pan.baidu.com/s/1ez7FaE2qy9m40JSvJr-VWA?pwd=x7fd

GitHub地址: https://github.com/jianchang512/pyvideotrans/releases/download/v3.66/win-videotrans-v3.67-tiny.7z

 

## v3.67 补丁包190MB

> 如果已安装过3.x版本，可下载补丁包后解压在sp.exe所在目录，覆盖已有sp.exe和文件夹

百度网盘下载地址: https://pan.baidu.com/s/1QnKCu4YO9OFxN-C7ONfTqA?pwd=c1mc

GitHub地址: https://github.com/jianchang512/pyvideotrans/releases/download/v3.66/win-PatchUpdate-3.67.7z


> 若补丁覆盖后打开提示`需下载完整包`，请升级显卡驱动、升级cuda到12.x、升级cudnn到cudnn9，并重新下载完整包


----

### 所有模型下载地址

为避免压缩包体积过大，预打包版只内置最小模型 tiny，识别效果不佳，效果更好的模型请点击下载 ，建议至少使用medium模型，推荐large-v2

https://github.com/jianchang512/stt/releases/tag/0.0