更新

1. 更新对新版modelscope FunASR API的支持。现在重新支持全自动中文标注。

Update

1. Update added support for the new version of ModelScope FunASR API. Full automatic Chinese annotation is now supported again.

更新

1. 更新对新版modelscope API的支持。
2. 增加一个对文本增加标点符号的方法。使用subfix punctuation --load_list demo.list命令对list文件进行标点。

Update

1. Enhanced support for the new version of the ModelScope API.
2. Added a method to add punctuation to text. Use the command subfix punctuation --load_list demo.list to apply punctuation to the list file.

更新

1. 现在webui --force_delete选项将默认为True，现在默认情况下在webui界面删除音频的时候，硬盘上的音频文件将同步删除。
2. 将.py文件版本的subfix_webui.py添加到主分支根目录。同时进行了更新，现在可以通过命令行在一个文件中进行中英文切换，同时支持对硬盘上的音频文件进行同步修改。支持指定端口启动。

Update

1. The webui --force_delete option will now default to True. When deleting audio files in the webui interface by default, the corresponding audio files on the hard drive will be synchronized and deleted.
2. Added the .py file version of subfix_webui.py to the root directory of the main branch. It has been updated to allow seamless switching between Chinese and English in a single file through the command line. Additionally, it now supports synchronized modifications to audio files on the hard drive. Support launch with a specified port.

更新

1. 现在webui --force_delete选项将默认为True，现在默认情况下在webui界面删除音频的时候，硬盘上的音频文件将同步删除。
2. 将.py文件版本的subfix_webui.py添加到主分支根目录。同时进行了更新，现在可以通过命令行在一个文件中进行中英文切换，同时支持对硬盘上的音频文件进行同步修改。

Update

1. The webui --force_delete option will now default to True. When deleting audio files in the webui interface by default, the corresponding audio files on the hard drive will be synchronized and deleted.
2. Added the .py file version of subfix_webui.py to the root directory of the main branch. It has been updated to allow seamless switching between Chinese and English in a single file through the command line. Additionally, it now supports synchronized modifications to audio files on the hard drive.

更新

1. 现在webui --force_delete选项将默认为True，现在默认情况下在webui界面删除音频的时候，硬盘上的音频文件将同步删除。

Update

1. Now the webui --force_delete option will default to True. By default, when deleting audio files through the web UI interface, the audio files on the hard drive will be synchronously deleted.

更新

1. 在requirements.txt中增加openai-whisper依赖。

Update

1. add openai-whisper in requirements.txt

更新

1. 修改部分文件，通过README中的方式，可以在Windows下使用自动标注功能。

Update

1. Modify certain files so that the automatic annotation function can be used on Windows, following the method described in the README.

更新

1. 在此版本开始，subfix将作为一个工具，可以在任意路径下，通过subfix命令来启动。
2. 支持modelscope多语言自动化标注。
3. 支持openai-whisper多语言自动化标注。
4. 增加说话人日志功能，进行数据预处理。通过subfix diarization命令可以提取出每个音频中说话次数最多的n个说话人。
5. 旧版请访问v1.2

Update

1. From this version onwards, subfix will serve as a tool that can be launched from any path using the subfix command.
2. Support for modelscope multi-language automatic annotation is now available.
3. Support for openai-whisper multi-language automatic annotation is now available.
4. Added speaker log functionality for data preprocessing. The subfix diarization command allows for the extraction of the top n speakers with the most speaking occurrences in each audio file.
5. For the previous version, please visit v1.2

This version is an archive of the version released as a separate file.

In later versions, this tool will be used as a module.

此版本为单独文件发布的版本的存档。

在之后版本中，本工具将作为一个模块来使用。