Johnson-Lin commited on
Commit
5d1701f
·
verified ·
1 Parent(s): c297058

Correct Tiny backbone parameter description to 7.9B-A1.3B

Browse files

Update the subtitle, introduction and parameter table. Preserve model assets and conversion provenance.

Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -22,7 +22,7 @@ tags:
22
  <div align="center">
23
  <img src="./figures/ant-bailing.png" width="60" alt="Ming-3.0-tts Logo" />
24
  <h1>Ming-3.0-tts-tiny</h1>
25
- <p><i>8B-A1B multilingual MoE speech synthesis with paralinguistic and free-form control.</i></p>
26
  </div>
27
 
28
  <p align="center">
@@ -31,7 +31,7 @@ tags:
31
  <a href="https://modelscope.cn/models/inclusionAI/Ming-3.0-tts-tiny">ModelScope</a>
32
  </p>
33
 
34
- This repository contains **Ming-3.0-tts-tiny**, the Ming-3.0-tts mixture-of-experts checkpoint built on Ling V3. It has 8B total parameters and activates about 1B parameters per token. The repository includes sharded weights, supporting model code, audio tokenizer, speaker encoder and text tokenizer.
35
 
36
  > **Inference requires the [Ming-3.0-tts GitHub repository](https://github.com/inclusionAI/Ming-3.0-tts).** This Hugging Face repository is for checkpoint distribution and does not provide a standalone inference setup. Clone the GitHub repository and follow its README for the current dependencies, inference code and usage instructions. The CLI scripts and Python APIs referenced below are provided by the GitHub repository.
37
 
@@ -40,7 +40,7 @@ This repository contains **Ming-3.0-tts-tiny**, the Ming-3.0-tts mixture-of-expe
40
  | Item | Value |
41
  | --- | --- |
42
  | Architecture | Mixture of Experts, Ling V3 backbone |
43
- | Parameters | 8B total, about 1B active |
44
  | Languages | Chinese, English, German, Spanish, French, Korean, Japanese |
45
  | Dialects | Sichuanese, Cantonese |
46
  | Python API | `ming_tiny.MingTinyTTS` |
 
22
  <div align="center">
23
  <img src="./figures/ant-bailing.png" width="60" alt="Ming-3.0-tts Logo" />
24
  <h1>Ming-3.0-tts-tiny</h1>
25
+ <p><i>7.9B-A1.3B multilingual MoE speech synthesis with paralinguistic and free-form control.</i></p>
26
  </div>
27
 
28
  <p align="center">
 
31
  <a href="https://modelscope.cn/models/inclusionAI/Ming-3.0-tts-tiny">ModelScope</a>
32
  </p>
33
 
34
+ This repository contains **Ming-3.0-tts-tiny**, the Ming-3.0-tts mixture-of-experts checkpoint built on Ling V3. Its Ling V3 backbone has 7.9B total parameters and activates about 1.3B parameters per token. The repository includes sharded weights, supporting model code, audio tokenizer, speaker encoder and text tokenizer.
35
 
36
  > **Inference requires the [Ming-3.0-tts GitHub repository](https://github.com/inclusionAI/Ming-3.0-tts).** This Hugging Face repository is for checkpoint distribution and does not provide a standalone inference setup. Clone the GitHub repository and follow its README for the current dependencies, inference code and usage instructions. The CLI scripts and Python APIs referenced below are provided by the GitHub repository.
37
 
 
40
  | Item | Value |
41
  | --- | --- |
42
  | Architecture | Mixture of Experts, Ling V3 backbone |
43
+ | Backbone parameters | 7.9B total, about 1.3B active per token |
44
  | Languages | Chinese, English, German, Spanish, French, Korean, Japanese |
45
  | Dialects | Sichuanese, Cantonese |
46
  | Python API | `ming_tiny.MingTinyTTS` |