v1.8.0

Release Notes

🔧 Improvements:

• Refined and optimized the overall code structure for better performance and maintainability.
• Enhanced the prompt to be more concise and applicable to a wider range of models.
• Improved fuzzy and precise matching in translation processes.

🐛 Bug Fixes:

• Fixed several errors occurring during the FFmpeg compression process.
• Resolved an issue with phrase errors caused by lack of initialization and null returns from model translations.

📝 Updates:

• Demucs vocal separation is no longer performed by default before transcription, addressing the issue of missing sentences and improving processing speed.
• Removed support for the whisperX replicate API to simplify the project as an open-source initiative.
• Adjusted the translation process to handle smaller segments, reducing the likelihood of errors.

---

发布说明

🔧 改进：

• 精简优化了整体代码结构，提高了性能和可维护性。
• 优化了提示词，使其更加精简，适用于更多模型。
• 改进了翻译过程中的模糊和精确匹配。

🐛 问题修复：

• 修复了在 FFmpeg 压制过程中发生的一些错误。
• 解决了由于未初始化和模型翻译返回空值导致的 phrase 错误。

📝 更新：

• 默认不在转录前进行 Demucs 人声分离，以解决遗漏句子的问题并提高处理速度。
• 删除了 whisperX replicate API 的支持，以简化开源项目。
• 调整翻译过程为处理更小的块，减少错误的可能性。

v1.7.1

Release Notes

🚀 New Features:

• Added MPS support for Demucs to improve performance.
• Implemented an error retry mechanism for batch processing.
• Set the default to use the Gemini model and updated documentation accordingly.
• Added an auto-update feature for ytdlp.

🐛 Bug Fixes:

• Corrected a long video segmentation error.
• Fixed loading issues with the local Chinese Whisper model.
• Improved audio splitting robustness and encoding handling.
• Resolved issues with handling reference audio prerequisites for GPT-SoVITS batch processing.
• Correctly implemented retry on translation failures.

🔧 Improvements:

• Updated the CPU-specific torch version in the installation process.
• Refactored to simplify the prompt reasoning chain due to minimal improvement.

📝 Updates:

• Removed the install option in the OneKey batch script.
• Added a section for the SaaS website in the documentation.

🚀 新功能：

• 为 Demucs 增加了 MPS 支持以提升性能。
• 实现了批处理的错误重试机制。
• 设置默认使用 Gemini 模型并相应更新了文档。
• 新增 ytdlp 自动更新功能。

🐛 问题修复：

• 修正了长视频分段错误。
• 修复了本地中文 Whisper 模型的加载问题。
• 改进了音频分割的鲁棒性和编码处理。
• 解决了 GPT-SoVITS 批处理的参考音频前提条件问题。
• 正确实现了翻译失败时的重试机制。

🔧 改进：

• 更新了安装过程中与 CPU 相关的 torch 版本。
• 重构以简化提示推理链条，因其改进效果有限。

📝 更新：

• 移除了 OneKey 批处理脚本中的安装选项。
• 在文档中添加了 SaaS 网站的部分。

v1.7.0

🚀 New Features:

• Enabled GPU acceleration for FFmpeg encoding (5x speed boost)
• Replaced UVR with Demucs for vocal isolation (5x speed boost)
• Automated torch version selection based on GPU in install.py

🐛 Bug Fixes:

• Resolved local whisperX video segmentation issue
• Fixed support for uppercase file extensions

🔧 Improvements:

• Simplified code structure
• Improved spacing between Chinese and English in subtitles with autocorrect
• Streamlined PyPI sources to official and Tsinghua mirrors

📝 Updates:

• One-click package and free test key are no longer provided
• Commercial SaaS version will be released tomorrow

🚀 新功能：

• 为FFmpeg编码启用GPU加速（5倍速度提升）
• 用Demucs替换UVR进行人声分离（5倍速度提升）
• 在install.py中根据GPU自动选择torch版本

🐛 问题修复：

• 解决本地whisperX视频分段问题
• 修复对大写文件扩展名的支持

🔧 改进：

• 简化代码结构
• 使用自动纠正改善字幕中中英文之间的间距
• 精简PyPI源为官方和清华镜像

📝 更新：

• 不再提供一键包和免费的测试key
• 商业SaaS版本将于明天发布

v1.6.4

🚀 New Features:

• Added m4a file support
• Automated Chinese transcription model download
• Optimized multilingual PyPI source selection

🐛 Bug Fixes:

• Fixed uppercase file extension issue
• Adjusted summary length to 4k characters

📝 Updates:

• New logo and documentation improvements
• Model uploaded to Docker Hub

🚀 新功能：

• 支持m4a文件格式
• 自动下载中文转录模型
• 优化多语言PyPI源选择

🐛 问题修复：

• 解决文件扩展名大写问题
• 调整摘要长度为4k字符

📝 更新：

• 新logo和文档改进
• 模型上传至Docker Hub

v1.6.3

• 🌐 Simplified the implementation of multilingual support . Chinese users, please apply the localization patch according to the installation documentation 🇨🇳. Support for other languages is TODO 📝

• 📚 Updated the technical documentation on the official website

• 🌐 简化了多语言的实现方式，中文用户请根据安装文档打上汉化补丁，其他语言的支持TODO

• 📚 更新了官网的技术文档

v1.6.2

📢 Announcement: Starting from this version, the open-source edition will only receive stability updates. Our commercial SaaS version is coming soon, stay tuned!

🎉 New Features

• Added support for audio file uploads
• Added Docker support

🛠️ Stability Fixes

• Fixed upload-related bugs
• Fixed audio format processing issues
• Optimized configuration file import
• Fixed OpenAI TTS configuration issues
• Fixed and optimized ffmpeg-related issues
• Added int8 support for older GPUs
• Optimized pip source selection and Hugging Face mirror choice during installation

📢 公告：从这个版本开始，开源版本将只进行稳定性更新。我们的商业SaaS版本即将推出，敬请期待！

🎉 新功能

• 添加对音频文件上传的支持
• 添加 Docker 支持

🛠️ 稳定性修复

• 修复上传相关 bug
• 修复音频格式处理问题
• 优化配置文件导入
• 修复 OpenAI TTS 配置问题
• 修复并优化 ffmpeg 相关问题
• 为旧款 GPU 添加 int8 支持
• 安装时优选pip源和选择huggingface镜像

v1.6.1

Enhance System Stability

• Revert FFmpeg installation due to encountered issues
• Resolve SoVITS configuration import problems
• Address YAML installation errors
• Replace Librosa with FFmpeg to mitigate compatibility issues
• Boost stability in batch processing mode"

提升系统稳定性

• 由于出现问题，回退 FFmpeg 安装
• 解决 SoVITS 配置导入问题
• 修复 YAML 安装错误
• 用 FFmpeg 替换 Librosa 以解决兼容性问题
• 增强批处理模式的稳定性

v1.6

• Refactored config.py into config.yaml, with corresponding restructuring of the codebase. The UI no longer requires clicking a save button.

• Automatic source selection during local installation

• Fixed an issue where language settings in batch mode were ineffective

• Improved stability for accessing Replicate

• 将 config.py 重构为 config.yaml ，代码库也跟随着进行了重构，UI中现在不需要点击保存了。

• 本地安装时自动选择源

• 修复了 batch mode 中填写语言无效的问题

• 现在访问replicate的方法更稳定了

v1.5.1

• Emergency fix for the blank nan bug and zh check bug in batch mode

• Fixed an issue where the language set in the tasks settings of version 1.5 was ineffective.

• Added mirror source selection for installation steps

• 紧急修复了 batch 模式的留空 nan bug和 zh 检查 bug，

• 修复了1.5版本tasks setting中设置的语言无效的问题

• 增加了安装步骤的镜像源选择

v1.5

Major Updates:

• Added batch processing functionality
• Simplified JSON key structure in prompts

Minor Improvements:

• Support for running in 6GB gpu memory environments
• Improved empty line detection in NLP step
• Fixed font issues in Linux environments
• Handled phrase alignment errors and prompted to retry UVR
• Limited file upload size to 500MB in Streamlit
• Made video compression a separate optional step
• Restructured README documentation

主要更新：

• 新增批量处理功能
• 简化promot中的JSON key结构

小改进：

• 改善了6G显存环境下的运行速度
• 改进了NLP步骤中的空行检测
• 修复Linux环境下的字体问题
• 处理短语对齐错误并提示重试UVR
• 在Streamlit中限制文件上传大小为500MB
• 是否压制视频单独作为可选项
• 重构了README文档