Hugging Face's logo

mispeech
/
Dasheng-AudioGen-Multilingual

Text-to-Audio
Transformers
Safetensors
dasheng_audiogen
feature-extraction
audio-generation
text-to-speech
text-to-music
sound-effects
diffusion
custom_code
Model card Files
xet
 Community
Jiahao mei commited on
Commit
dfc0841
·
verified ·
1 Parent(s): 476ea22

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. README.md +13 -1
  2. README_zh.md +13 -1
README.md CHANGED
@@ -27,7 +27,7 @@ pipeline_tag: text-to-audio
27
 
28
  **Dasheng-AudioGen-Multilingual** is the multilingual variant of [Dasheng-AudioGen](https://huggingface.co/mispeech/Dasheng-AudioGen). It replaces the text encoder with `google/mt5-large`, enabling text-to-audio generation from prompts in multiple languages.
29
 
30
- - GitHub: [https://github.com/nicollasm/Dasheng-AudioGen](https://github.com/nicollasm/Dasheng-AudioGen)
31
  - Demo: [https://huggingface.co/spaces/mispeech/Dasheng-AudioGen](https://huggingface.co/spaces/mispeech/Dasheng-AudioGen)
32
  - Web Demo: [https://nieeim.github.io/Dasheng-AudioGen-Web/](https://nieeim.github.io/Dasheng-AudioGen-Web/)
33
  - Base model: [mispeech/Dasheng-AudioGen](https://huggingface.co/mispeech/Dasheng-AudioGen)
@@ -66,6 +66,18 @@ Training data language distribution:
66
 
67
  ## Usage
68
 
 
 
 
 
 
 
 
 
 
 
 
 
69
  ### Python API
70
 
71
  ```python
 
27
 
28
  **Dasheng-AudioGen-Multilingual** is the multilingual variant of [Dasheng-AudioGen](https://huggingface.co/mispeech/Dasheng-AudioGen). It replaces the text encoder with `google/mt5-large`, enabling text-to-audio generation from prompts in multiple languages.
29
 
30
+ - GitHub: [https://github.com/xiaomi-research/dasheng-audiogen](https://github.com/xiaomi-research/dasheng-audiogen)
31
  - Demo: [https://huggingface.co/spaces/mispeech/Dasheng-AudioGen](https://huggingface.co/spaces/mispeech/Dasheng-AudioGen)
32
  - Web Demo: [https://nieeim.github.io/Dasheng-AudioGen-Web/](https://nieeim.github.io/Dasheng-AudioGen-Web/)
33
  - Base model: [mispeech/Dasheng-AudioGen](https://huggingface.co/mispeech/Dasheng-AudioGen)
 
66
 
67
  ## Usage
68
 
69
+ ### Installation
70
+
71
+ ```bash
72
+ git clone https://github.com/xiaomi-research/dasheng-audiogen.git
73
+ cd dasheng-audiogen
74
+ conda create -n dasheng-audiogen python=3.10
75
+ conda activate dasheng-audiogen
76
+ pip install -r requirements.txt
77
+ ```
78
+
79
+ > torch 2.8.0+cu128 is recommended.
80
+
81
  ### Python API
82
 
83
  ```python
README_zh.md CHANGED
@@ -4,7 +4,7 @@
4
 
5
  **Dasheng-AudioGen-Multilingual** 是 [Dasheng-AudioGen](https://huggingface.co/mispeech/Dasheng-AudioGen) 的多语言版本。它将文本编码器替换为 `google/mt5-large`，支持使用多种语言的 prompt 进行音频生成。
6
 
7
- - GitHub: [https://github.com/nicollasm/Dasheng-AudioGen](https://github.com/nicollasm/Dasheng-AudioGen)
8
  - Demo: [https://huggingface.co/spaces/mispeech/Dasheng-AudioGen](https://huggingface.co/spaces/mispeech/Dasheng-AudioGen)
9
  - Web Demo: [https://nieeim.github.io/Dasheng-AudioGen-Web/](https://nieeim.github.io/Dasheng-AudioGen-Web/)
10
  - 基础模型: [mispeech/Dasheng-AudioGen](https://huggingface.co/mispeech/Dasheng-AudioGen)
@@ -43,6 +43,18 @@
43
 
44
  ## 使用方法
45
 
 
 
 
 
 
 
 
 
 
 
 
 
46
  ### Python API
47
 
48
  ```python
 
4
 
5
  **Dasheng-AudioGen-Multilingual** 是 [Dasheng-AudioGen](https://huggingface.co/mispeech/Dasheng-AudioGen) 的多语言版本。它将文本编码器替换为 `google/mt5-large`，支持使用多种语言的 prompt 进行音频生成。
6
 
7
+ - GitHub: [https://github.com/xiaomi-research/dasheng-audiogen](https://github.com/xiaomi-research/dasheng-audiogen)
8
  - Demo: [https://huggingface.co/spaces/mispeech/Dasheng-AudioGen](https://huggingface.co/spaces/mispeech/Dasheng-AudioGen)
9
  - Web Demo: [https://nieeim.github.io/Dasheng-AudioGen-Web/](https://nieeim.github.io/Dasheng-AudioGen-Web/)
10
  - 基础模型: [mispeech/Dasheng-AudioGen](https://huggingface.co/mispeech/Dasheng-AudioGen)
 
43
 
44
  ## 使用方法
45
 
46
+ ### 安装
47
+
48
+ ```bash
49
+ git clone https://github.com/xiaomi-research/dasheng-audiogen.git
50
+ cd dasheng-audiogen
51
+ conda create -n dasheng-audiogen python=3.10
52
+ conda activate dasheng-audiogen
53
+ pip install -r requirements.txt
54
+ ```
55
+
56
+ > 推荐使用 torch 2.8.0+cu128。
57
+
58
  ### Python API
59
 
60
  ```python