Starstruk Logo MarkStarstruk

v3.40

jianchang512/pyvideotransv3.40Dec 21, 2024by jianchang512

AI Summary

Users can now customize the three-step reflection translation prompt, and the audio acceleration and slow-motion algorithms have been optimized. OpenAI TTS compatibility with third-party TTS is also added.

Key Highlights

  • Customizable three-step reflection translation prompt
  • Optimized audio acceleration and slow-motion algorithms
  • OpenAI TTS compatibility with third-party TTS
  • MacOS/Linux upgrade instructions provided

New Features

  • Customizable reflection prompt
  • OpenAI TTS compatibility

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` 
>  `pip3 install --no-deps  --force-reinstall "faster-whisper @ https://github.com/SYSTRAN/faster-whisper/archive/refs/heads/master.tar.gz"`

## ChangeLogs

- 三步反思翻译提示词可在`videotrans/prompts/srt/fansi3.txt`中修改
- 优化音频加速、视频慢速算法
- openai tts兼容第三方tts


## Win v3.41 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/1_XgiqYqqBElO_40pCmgdhA?pwd=gp6h

GitHub下载地址(仅含tiny模型):  https://github.com/jianchang512/pyvideotrans/releases/download/v3.40/win-videotrans-v3.41-tiny.7z

 

## v3.41 补丁包190MB

> 如果已安装过2.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/1zmaJuxRulf7zQBFLJaffNA?pwd=e6bs

GitHub地址: https://github.com/jianchang512/pyvideotrans/releases/download/v3.40/win-PatchUpdate-3.41.7z


> 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，然后再使用该补丁包覆盖。
> 
> 或下载该压缩包 https://pan.baidu.com/s/1LOjr9ampzqp7eBrjNgSTHQ?pwd=wrkn   解压后(也可从此地址下载  
 https://github.com/jianchang512/pyvideotrans/releases/download/v3.15/cuda-torch-pathupdate.7z  )，复制其内的 `_internal`文件夹覆盖软件目录内的`_internal`文件夹 


----

api.exe 下载，api.exe是用于通过http请求调用的文件，不可独立使用，需先下载完整包，然后下载api.exe和 sp.exe放在同一目录下
 
api.exe下载地址：https://github.com/jianchang512/pyvideotrans/releases/download/v3.36/api.exe


----

### 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