Skip to content

SOTA-Class TTS at Compact Scale

Python

Simplify architecture attribution

rootac499ac7f8f5
diff --git a/NOTICE b/NOTICE
index 291bf91..b8596dc 100644
--- a/NOTICE
+++ b/NOTICE
@@ -1,8 +1,4 @@
audio8_tts
-The audio8_tts model weights were trained from scratch and do not contain
-weights from Fish Audio models.
-
The model architecture is inspired by the DualAR design used in Fish Audio
-S2 Pro. Fish Audio and Fish S2 Pro are trademarks or product names of their
-respective owners. This attribution does not imply endorsement.
+S2 Pro.
diff --git a/README.md b/README.md
index 6b0e97a..6dc196f 100644
--- a/README.md
+++ b/README.md
@@ -33,8 +33,7 @@ The Preview checkpoint performs best in the following languages:
## Architecture
audio8_tts uses a DualAR architecture inspired by
-[Fish Audio S2 Pro](https://github.com/fishaudio/fish-speech). The model weights
-were trained from scratch and do not use Fish Audio weights.
+[Fish Audio S2 Pro](https://github.com/fishaudio/fish-speech).
| Component | Configuration |
|---|---|
@@ -218,5 +217,4 @@ Code and model weights in this repository are released under the
[Apache License 2.0](LICENSE). See [NOTICE](NOTICE) for attribution details.
We thank the Fish Audio team for publishing the DualAR architecture used in
-Fish S2 Pro. audio8_tts adopts the architectural idea while using independently
-trained weights.
+Fish S2 Pro.
diff --git a/README_zh.md b/README_zh.md
index fa0e22f..d6f0b07 100644
--- a/README_zh.md
+++ b/README_zh.md
@@ -29,7 +29,7 @@ Preview 模型当前表现较好、推荐使用的语言如下:
## 模型结构
audio8_tts 沿用了 [Fish Audio S2 Pro](https://github.com/fishaudio/fish-speech)
-的 DualAR 架构思路。模型权重完全从零训练,没有使用 Fish Audio 权重。
+的 DualAR 架构思路。
| 组件 | 配置 |
|---|---|
@@ -196,5 +196,4 @@ Fish S2 Pro 因官方评测使用自身 normalizer 而重新评测,Higgs Audio
本仓库代码和模型权重采用 [Apache License 2.0](LICENSE),归属说明见 [NOTICE](NOTICE)。
-感谢 Fish Audio 团队公开 Fish S2 Pro 使用的 DualAR 架构。audio8_tts 借鉴了该架构思路,模型权重
-则为独立从零训练。
+感谢 Fish Audio 团队公开 Fish S2 Pro 使用的 DualAR 架构。