Skip to content

Support indextts-2 #265

@zhzLuke96

Description

@zhzLuke96

https://github.com/index-tts/index-tts
https://arxiv.org/abs/2506.21619
https://index-tts.github.io/index-tts2.github.io/

特点:

  • base model 是 qwen3
  • 情绪特征分离

目前还没开源

问题:
情绪特征提取不只是从文本,也可以从音频中提取,这块目前还没支持,得增加类似音色提取的管线处理这个逻辑。
(不过,情绪提取的效果怎么样还不确定,也许和直接1shot没区别?)

Metadata

Metadata

Assignees

No one assigned

    Labels

    StoryNext iteration summary and TODO listdelayedThis issue is currently difficult to troubleshoot and has been deferred for future resolution

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions