Update README.md

README.md

@@ -1,3 +1,84 @@
 ---
 license: openrail
+
+datasets:
+- LinkSoul/LLaSA-Audio-Instructions
+language:
+- zh
+- en
 ---
+
+
+# LLaSA: Large Language and Speech Assistant
+
+开源，可商用的**中英文双语语音-语言助手 LLaSA 以及中英文语音 SFT 数据集 LLaSA-Audio-Instructions**，第一个支持中英文语音-文本多模态对话的开源可商用对话模型。
+
+<!--
+<div align="center">
+    <img src="https://huggingface.co/LinkSoul/LLaSA-Cllama2/blob/main/meta/preview.jpg" width="40%">
+</div>
+
+![LLaSA](meta/preview.jpg)
+--> 
+
+## 基础演示
+
+![Base Demo](meta/demo.gif)
+
+## 在线试玩
+
+> Talk is cheap, Show you the Demo.
+- [Demo 地址 / HuggingFace Spaces](https://huggingface.co/spaces/LinkSoul/LLaSA) 
+
+## 资源下载
+
+- 模型：
+  - [LLaSA-Chinese-Llama-2-7B](https://huggingface.co/LinkSoul/LLaSA-Cllama2)
+  - [LLaSA-Baichuan-7B](https://huggingface.co/LinkSoul/LLaSA-Baichuan)
+
+- 数据集：[LLaSA-Audio-Instructions](https://huggingface.co/datasets/LinkSoul/LLaSA-Audio-Instructions)
+
+## 环境安装
+```shell
+# clone the repository
+git clone https://github.com/LinkSoul-AI/LLaSA
+cd LLaSA
+
+# install package
+conda create -n llasa python=3.10 -y
+conda activate llasa
+pip install --upgrade pip
+pip install -e .
+```
+
+## 快速测试
+
+```shell
+export LLASA_DEVICE="cuda:0"
+python infer.py \
+    --input_audio_file PATH/TO/YOUR/AUDIO \
+    --llasa_model PATH/TO/LLaSA/MODEL \
+    --llasa_audio_tower PATH/TO/WHISPER/MODEL \
+    --llm_type "Chinese_llama2" or "baichuan" \
+```
+
+## TODO
+- 如何训练
+- int4 量化
+- docker 部署
+
+## 相关项目
+- [Chinese-Llama-2-7B](https://huggingface.co/LinkSoul/Chinese-Llama-2-7b)
+- [Whisper](https://ai.meta.com/llama/)
+- [baichuan-inc/Baichuan-7B](https://huggingface.co/baichuan-inc/Baichuan-7B)
+
+
+## 项目协议
+
+[Apache-2.0 license](https://github.com/LinkSoul-AI/LLaSA/blob/main/LICENSE)
+
+## 微信交流群
+<!--
+<img src="meta/QRcode.jpg" alt="微信交流群" width="300"/>
+-->
+欢迎加入[微信群](meta/QRcode.jpg)