License clarification for emotion2vec+ base weights (paid desktop app, ONNX export downloaded on first use)

未关闭
#3,776 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 1 天内回复

还没有人认领这个 Issue。

评估

难度
5/5
预计耗时
一周以上
新手友好度
20/100
Issue 类型
文档
描述清晰度
描述清楚
活跃度
活跃
技术栈
python, pytorch
领域
documentation

调研方向

Start by checking the MODEL_LICENSE version and the emotion2vec model card for the named upstream weights and ONNX export. The issue asks maintainers to clarify commercial use, attribution, the meaning of the stated restriction, and whether later licence changes affect these revisions; a complete resolution should answer those questions or point to the responsible authors.

由索引模型根据 Issue 内容生成。

描述

Hello FunASR maintainers, this is a licensing question only — no technical support needed.

Artefacts

Upstream weights emotion2vec/emotion2vec_plus_base (Hugging Face), revision b318240bfe67db81a8c572ecb37ce9c3759b81c9
model.pt SHA-256 60710b5aae1dbe69bdac8920028fb05882d4314fd09031922b4b61ee9e7aadbd
ONNX export used pankotaro/emotion2vec-plus-base-onnx, revision 334d4376993c381f059cc36d2c6ed05cbfec5ca1
emotion2vec_plus_base.onnx SHA-256 a9ddcc2f4254c3da1dc2a51ea5e38df32e2a84724519e343499f05f0ea2b3506
emotion2vec_head.json SHA-256 69666ff0d2873f016f5a8187b685e32a2c419130876bf97b069c981c44cfe02c

Why the metadata does not settle it: the model card says license: other / model-license, and license_link points to the FunASR repository home page, not to a fixed licence file. The current MODEL_LICENSE v1.1 allows use, modification and sharing with attribution, but does not mention commercial use, calls the software "provided for reference and learning purposes only", and may be revised.

Intended use: local speech emotion recognition inside a paid, proprietary desktop application (one-time purchase or subscription). Weights unmodified and not bundled: the app downloads the ONNX file on first use and verifies its SHA-256. No hosted API, no standalone sale.

Questions

  1. Which licence and version apply to these files?
  2. Is this commercial use permitted?
  3. Is "reference and learning purposes only" a restriction or a disclaimer?
  4. What attribution wording do you require?
  5. Would future licence revisions apply to these fixed revisions?

If the decision rests with the emotion2vec authors (https://github.com/ddlBoJack/emotion2vec), a pointer would be appreciated. Thank you!

主要语言
Python
星标
21k
派生
2.1k
平均合并
15 小时 16 分钟
30 天内合并 PR
85

环境准备

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

modelscope/FunASR 的其他 Issue

查看 modelscope/FunASR 的全部 Issue

相似的 Issue

更多 Python Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。