Update README.md (#1)
Browse files- Update README.md (1ecc90bad9207f97244138a07c002290f26f66a7)
Co-authored-by: mrh <[email protected]>
README.md
CHANGED
|
@@ -4,14 +4,9 @@ license: apache-2.0
|
|
| 4 |
# Step-Audio-Tokenizer
|
| 5 |
|
| 6 |
|
| 7 |
-
Step-Audio LLM是业界首个拥有1300亿参数的类人化统一端到端模型，整合了多模态语音理解与生成能力，涵盖歌声合成、工具调用、角色扮演及多语言/方言理解与合成等功能。
|
| 8 |
-
|
| 9 |
Step-Audio LLM is the industry’s first 130-billion parameter hu-manlike unified end-to-end model that integrates multimodal speech un-derstanding and generation capabilities, including singing voice synthesis, tool utilization, role-play and multilingual/dialectal comprehension and synthesis.
|
| 10 |
|
| 11 |
-
本仓库提供 Step-Audio LLM的speech tokenizer模块。针对linguistic tokenization，我们采用 Paraformer 编码器的输出特征，将其量化至离散表示，码率为 16.7 Hz；针对semantic tokenization，我们使用 CosyVoice 的tokenizer——专为高效编码、自然且富有表现力的语音输出设计，码率为 25 Hz。
|
| 12 |
-
|
| 13 |
This repository provides the speech tokenizer component of Step-Audio LLM. For linguistic tokenization, we utilize the output from the Paraformer encoder, which is quantized into discrete representations at a token rate of 16.7 Hz. For semantic tokenization, we employ CosyVoice’s tokenizer, specifically designed to efficiently encode features essential for generating natural and expressive speech outputs, operating at a token rate of 25 Hz.
|
| 14 |
|
| 15 |
-
|
| 16 |
-
|
| 17 |
For more information, please refer to our repository: [Step-Audio](https://github.com/stepfun-ai/Step-Audio).
|
|
|
|
| 4 |
# Step-Audio-Tokenizer
|
| 5 |
|
| 6 |
|
|
|
|
|
|
|
| 7 |
Step-Audio LLM is the industry’s first 130-billion parameter hu-manlike unified end-to-end model that integrates multimodal speech un-derstanding and generation capabilities, including singing voice synthesis, tool utilization, role-play and multilingual/dialectal comprehension and synthesis.
|
| 8 |
|
|
|
|
|
|
|
| 9 |
This repository provides the speech tokenizer component of Step-Audio LLM. For linguistic tokenization, we utilize the output from the Paraformer encoder, which is quantized into discrete representations at a token rate of 16.7 Hz. For semantic tokenization, we employ CosyVoice’s tokenizer, specifically designed to efficiently encode features essential for generating natural and expressive speech outputs, operating at a token rate of 25 Hz.
|
| 10 |
|
| 11 |
+
## More information
|
|
|
|
| 12 |
For more information, please refer to our repository: [Step-Audio](https://github.com/stepfun-ai/Step-Audio).
|