v4.06

X-PLUG/mPLUG-DocOwlv4.06Jul 18, 2026by jianchang512

AI Summary

This update focuses on localization features, integrating ConfuciusTTS and OmniVoice as built-in voice channels and upgrading chatterbox-tts to v3.0. It also adds support for speaker identification in local SRT files for better role assignment during video translation.

Key Highlights

  • ConfuciusTTS and OmniVoice added as built-in voice channels
  • chatterbox-tts upgraded to version 3.0
  • Support for speaker flags (e.g., [spk1]) in local SRT files
  • Optimized audio acceleration and slow-motion algorithms
  • Optimized startup speed

New Features

  • Built-in voice channels
  • Speaker flag support in SRT files
  • Audio/video algorithm optimizations

Full Release Notes

**v4.06更新日志**

- 配音渠道 ConfuciusTTS / OmniVoice 改为内置
- 优化音频加速、视频慢速算法
- 升级 chatterbox-tts 到v3.0
- 优化启动速度
- 使用本地已有srt字幕文件时支持带有说话人标志
> 如果每行字幕均以`[spk数字]`或`[speaker数字]`开头,例如`[spk1]`/`[speaker2]`等开头,则将此识别为说话人,在单视频模式下,可在字幕翻译完成后,按说话人分配配音角色。批量视频翻译模式下忽略


----

### v4.06 完整包下载(2.7G)

百度网盘地址: https://pan.baidu.com/s/17HPQ4RKB7zCZibB6GwR5yg?pwd=1234

huggingface.co:  https://huggingface.co/mortimerme/repocollect/resolve/main/win-pyvideotrans-v4.06-0720.7z?download=true


### v4.06 补丁包下载(300MB)

> 补丁包不可单独使用,仅用于覆盖旧包, **覆盖后如果闪退,请下载完整包**
> 
> 如果补丁包覆盖后报错,建议下载完整包。

百度网盘地址: https://pan.baidu.com/s/1ilW4gn1Py8JYDQG5-Xn5Ig?pwd=1234

huggingface.co: https://huggingface.co/mortimerme/repocollect/resolve/main/win-PatchUpdate-v4.06-0720-2.7z?download=true





<details><summary>下载上一个版本(v4.03)</summary>

> 如果遇到新版bug无法使用,可重新下载上个版本

**v4.03 完整包下载(2.7G)**

- 百度网盘地址: https://pan.baidu.com/s/1yENT4t7E5K1OrbOFS7WEvA?pwd=1234

- huggingface.co:  https://huggingface.co/mortimerme/repocollect/resolve/main/win-pyvideotrans-v4.03-0626.7z?download=true


**v4.03 补丁包下载(260MB)**

- 百度网盘地址: https://pan.baidu.com/s/16kV6giJdip30l5uJe5XccA?pwd=1234

- huggingface.co: https://huggingface.co/mortimerme/repocollect/resolve/main/win-PatchUpdate-v4.03-0626.7z?download=true

</details>

## What's Changed
* fix(cli): support direct voice identifier in role list mapping by @minh-vt in https://github.com/jianchang512/pyvideotrans/pull/1156

## New Contributors
* @minh-vt made their first contribution in https://github.com/jianchang512/pyvideotrans/pull/1156

**Full Changelog**: https://github.com/jianchang512/pyvideotrans/compare/v4.05...v4.06