2024-06-02 22:25:19 +08:00
|
|
|
|
Real time interactive streaming digital human, realize audio video synchronous dialogue. It can basically achieve commercial effects.
|
|
|
|
|
实时交互流式数字人,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
|
|
|
|
|
2024-06-17 08:21:03 +08:00
|
|
|
|
[ernerf效果](https://www.bilibili.com/video/BV1PM4m1y7Q2/) [musetalk效果](https://www.bilibili.com/video/BV1gm421N7vQ/) [wav2lip效果](https://www.bilibili.com/video/BV1Bw4m1e74P/)
|
2023-12-19 09:41:52 +08:00
|
|
|
|
|
2024-10-05 17:25:01 +08:00
|
|
|
|
## 为避免与3d数字人混淆,原项目metahuman-stream改名为livetalking,原有链接地址继续可用
|
|
|
|
|
|
2024-03-31 12:01:28 +08:00
|
|
|
|
## Features
|
2024-06-17 08:21:03 +08:00
|
|
|
|
1. 支持多种数字人模型: ernerf、musetalk、wav2lip
|
2024-05-26 18:07:22 +08:00
|
|
|
|
2. 支持声音克隆
|
2024-07-04 09:43:56 +08:00
|
|
|
|
3. 支持数字人说话被打断
|
2024-03-31 12:01:28 +08:00
|
|
|
|
4. 支持全身视频拼接
|
2024-04-21 18:19:24 +08:00
|
|
|
|
5. 支持rtmp和webrtc
|
2024-05-04 10:10:41 +08:00
|
|
|
|
6. 支持视频编排:不说话时播放自定义视频
|
2024-03-31 12:01:28 +08:00
|
|
|
|
|
2024-01-13 20:12:08 +08:00
|
|
|
|
## 1. Installation
|
2023-12-19 09:41:52 +08:00
|
|
|
|
|
2024-01-13 20:12:08 +08:00
|
|
|
|
Tested on Ubuntu 20.04, Python3.10, Pytorch 1.12 and CUDA 11.3
|
2023-12-19 09:41:52 +08:00
|
|
|
|
|
2024-01-13 20:12:08 +08:00
|
|
|
|
### 1.1 Install dependency
|
2023-12-19 09:41:52 +08:00
|
|
|
|
|
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 activate nerfstream
|
2024-05-19 18:32:40 +08:00
|
|
|
|
conda install pytorch==1.12.1 torchvision==0.13.1 cudatoolkit=11.3 -c pytorch
|
2023-12-19 09:41:52 +08:00
|
|
|
|
pip install -r requirements.txt
|
2024-09-08 22:53:37 +08:00
|
|
|
|
#如果不训练ernerf模型,不需要安装下面的库
|
2023-12-19 09:41:52 +08:00
|
|
|
|
pip install "git+https://github.com/facebookresearch/pytorch3d.git"
|
2023-12-28 13:11:18 +08:00
|
|
|
|
pip install tensorflow-gpu==2.8.0
|
2024-05-19 18:32:40 +08:00
|
|
|
|
pip install --upgrade "protobuf<=3.20.1"
|
2023-12-19 09:41:52 +08:00
|
|
|
|
```
|
2024-08-23 23:19:38 +08:00
|
|
|
|
如果用pytorch2.1,torchvision用0.16(可以去torchvision官网根据pytorch版本找匹配的),cudatoolkit可以不用装
|
2024-05-19 18:32:40 +08:00
|
|
|
|
安装常见问题[FAQ](/assets/faq.md)
|
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
|
|
|
|
|
|
|
|
|
|
2024-05-02 20:32:28 +08:00
|
|
|
|
## 2. Quick Start
|
2024-06-02 22:25:19 +08:00
|
|
|
|
默认采用ernerf模型,webrtc推流到srs
|
2024-06-22 16:11:44 +08:00
|
|
|
|
### 2.1 运行srs
|
2023-12-19 09:41:52 +08:00
|
|
|
|
```
|
2024-05-02 20:32:28 +08:00
|
|
|
|
export CANDIDATE='<服务器外网ip>'
|
|
|
|
|
docker run --rm --env CANDIDATE=$CANDIDATE \
|
|
|
|
|
-p 1935:1935 -p 8080:8080 -p 1985:1985 -p 8000:8000/udp \
|
|
|
|
|
registry.cn-hangzhou.aliyuncs.com/ossrs/srs:5 \
|
|
|
|
|
objs/srs -c conf/rtc.conf
|
2023-12-19 09:41:52 +08:00
|
|
|
|
```
|
|
|
|
|
|
2024-01-13 20:12:08 +08:00
|
|
|
|
### 2.2 启动数字人:
|
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
|
|
|
|
```
|
|
|
|
|
|
2024-06-10 13:10:21 +08:00
|
|
|
|
用浏览器打开http://serverip:8010/rtcpushapi.html, 在文本框输入任意文字,提交。数字人播报该段文字
|
2024-05-02 20:32:28 +08:00
|
|
|
|
备注:服务端需要开放端口 tcp:8000,8010,1985; udp:8000
|
2024-01-28 07:58:46 +08:00
|
|
|
|
|
2024-03-31 12:01:28 +08:00
|
|
|
|
## 3. More Usage
|
2024-08-23 23:19:38 +08:00
|
|
|
|
使用说明: <https://livetalking-doc.readthedocs.io/>
|
2024-01-13 20:15:09 +08:00
|
|
|
|
|
2024-03-23 18:15:35 +08:00
|
|
|
|
## 4. Docker Run
|
2024-06-17 08:21:03 +08:00
|
|
|
|
不需要前面的安装,直接运行。
|
2024-01-07 14:19:21 +08:00
|
|
|
|
```
|
2024-06-22 16:11:44 +08:00
|
|
|
|
docker run --gpus all -it --network=host --rm registry.cn-beijing.aliyuncs.com/codewithgpu2/lipku-metahuman-stream:vjo1Y6NJ3N
|
2024-01-07 14:19:21 +08:00
|
|
|
|
```
|
2024-06-17 08:21:03 +08:00
|
|
|
|
代码在/root/metahuman-stream,先git pull拉一下最新代码,然后执行命令同第2、3步
|
2024-01-07 14:19:21 +08:00
|
|
|
|
|
2024-09-03 20:43:30 +08:00
|
|
|
|
提供如下镜像
|
|
|
|
|
- autodl镜像: <https://www.codewithgpu.com/i/lipku/metahuman-stream/base>
|
2024-05-30 11:39:43 +08:00
|
|
|
|
[autodl教程](autodl/README.md)
|
2023-12-19 09:41:52 +08:00
|
|
|
|
|
2024-06-09 09:04:04 +08:00
|
|
|
|
|
2024-08-23 23:19:38 +08:00
|
|
|
|
## 5. 性能分析
|
2024-01-13 20:12:08 +08:00
|
|
|
|
1. 帧率
|
2024-09-08 22:53:37 +08:00
|
|
|
|
在Tesla T4显卡上测试整体fps为18左右,如果去掉音视频编码推流,帧率在20左右。用4090显卡可以达到40多帧/秒。
|
2024-01-13 20:12:08 +08:00
|
|
|
|
2. 延时
|
2024-04-05 08:55:21 +08:00
|
|
|
|
整体延时3s左右
|
|
|
|
|
(1)tts延时1.7s左右,目前用的edgetts,需要将每句话转完后一次性输入,可以优化tts改成流式输入
|
|
|
|
|
(2)wav2vec延时0.4s,需要缓存18帧音频做计算
|
2024-05-02 20:32:28 +08:00
|
|
|
|
(3)srs转发延时,设置srs服务器减少缓冲延时。具体配置可看 https://ossrs.net/lts/zh-cn/docs/v5/doc/low-latency
|
2024-01-13 20:12:08 +08:00
|
|
|
|
|
2024-08-03 17:26:35 +08:00
|
|
|
|
|
2024-08-23 23:19:38 +08:00
|
|
|
|
## 6. TODO
|
2024-01-28 07:58:46 +08:00
|
|
|
|
- [x] 添加chatgpt实现数字人对话
|
2024-02-25 19:00:10 +08:00
|
|
|
|
- [x] 声音克隆
|
2024-05-04 10:10:41 +08:00
|
|
|
|
- [x] 数字人静音时用一段视频代替
|
2024-05-26 18:07:22 +08:00
|
|
|
|
- [x] MuseTalk
|
2024-06-17 08:21:03 +08:00
|
|
|
|
- [x] Wav2Lip
|
2024-08-03 17:26:35 +08:00
|
|
|
|
- [ ] TalkingGaussian
|
2023-12-19 09:41:52 +08:00
|
|
|
|
|
2024-08-23 23:19:38 +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.
|
|
|
|
|
* 知识星球: https://t.zsxq.com/7NMyO 沉淀高质量常见问题、最佳实践经验、问题解答
|
|
|
|
|
* 微信公众号:数字人技术
|
2024-06-02 22:25:19 +08:00
|
|
|
|
![](https://mmbiz.qpic.cn/sz_mmbiz_jpg/l3ZibgueFiaeyfaiaLZGuMGQXnhLWxibpJUS2gfs8Dje6JuMY8zu2tVyU9n8Zx1yaNncvKHBMibX0ocehoITy5qQEZg/640?wxfrom=12&tp=wxpic&usePicPrefetch=1&wx_fmt=jpeg&from=appmsg)
|
2024-06-09 09:04:04 +08:00
|
|
|
|
|