livetalking/README.md

82 lines
2.0 KiB
Markdown
Raw Normal View History

2023-12-28 13:11:18 +08:00
A streaming digital human based on the Ernerf model realize audio video synchronous dialogue. It can basically achieve commercial effects.
基于ernerf模型的流式数字人实现音视频同步对话。基本可以达到商用效果
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
## Installation
2023-12-19 09:41:52 +08:00
2024-01-06 21:17:34 +08:00
Tested on Ubuntu 18.04, Python3.10, Pytorch 1.12 and CUDA 11.3
2023-12-19 09:41:52 +08:00
### Install dependency
2023-12-28 13:11:18 +08:00
```bash
2024-01-06 21:17:34 +08:00
conda create -n nerfstream python=3.10
conda install pytorch==1.12.1 torchvision==0.13.1 cudatoolkit=11.3 -c pytorch
conda activate nerfstream
2023-12-19 09:41:52 +08:00
pip install -r requirements.txt
pip install "git+https://github.com/facebookresearch/pytorch3d.git"
2023-12-28 13:11:18 +08:00
pip install tensorflow-gpu==2.8.0
2023-12-19 09:41:52 +08:00
```
2023-12-28 13:11:18 +08:00
linux cuda环境搭建可以参考这篇文章 https://zhuanlan.zhihu.com/p/674972886
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
安装rtmpstream库
参照 https://github.com/lipku/python_rtmpstream
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
## Run
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
### 运行rtmpserver (srs)
2023-12-19 09:41:52 +08:00
```
2023-12-28 13:11:18 +08:00
docker run --rm -it -p 1935:1935 -p 1985:1985 -p 8080:8080 registry.cn-hangzhou.aliyuncs.com/ossrs/srs:5
2023-12-19 09:41:52 +08:00
```
2023-12-28 13:11:18 +08:00
### 启动:
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
```python
python app.py
2023-12-19 09:41:52 +08:00
```
2023-12-28 13:11:18 +08:00
如果访问不了huggingface在运行前
2023-12-19 09:41:52 +08:00
```
2023-12-28 13:11:18 +08:00
export HF_ENDPOINT=https://hf-mirror.com
2023-12-19 09:41:52 +08:00
```
2023-12-28 13:11:18 +08:00
运行成功后用vlc访问rtmp://serverip/live/livestream
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
### 网页端数字人播报输入文字
安装并启动nginx
2023-12-19 09:41:52 +08:00
```
2023-12-28 13:11:18 +08:00
apt install nginx
nginx
2023-12-19 09:41:52 +08:00
```
2023-12-28 13:11:18 +08:00
修改echo.html中websocket和视频播放地址将serverip替换成实际服务器ip
然后将echo.html和mpegts-1.7.3.min.js拷到/var/www/html下
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
启动数字人
2023-12-19 09:41:52 +08:00
```python
python app.py
```
2023-12-28 13:11:18 +08:00
用浏览器打开http://serverip/echo.html在文本框输入任意文字提交。数字人播报该段文字
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
## Data flow
![](/assets/dataflow.png)
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
## 数字人模型文件
可以替换成自己训练的模型(https://github.com/Fictionarry/ER-NeRF)
2023-12-19 09:41:52 +08:00
```python
.
├── data
2023-12-28 13:11:18 +08:00
│ ├── data_kf.json
2023-12-19 09:41:52 +08:00
│ ├── pretrained
│ └── └── ngp_kg.pth
```
2023-12-28 13:11:18 +08:00
## TODO
- 添加chatgpt实现数字人对话
- 声音克隆
- 数字人静音时用一段视频代替
2023-12-19 09:41:52 +08:00
2023-12-28 13:11:18 +08:00
如果本项目对你有帮助帮忙点个star。也欢迎感兴趣的朋友一起来完善该项目。
Email: lipku@foxmail.com