|
|
|
@ -7,6 +7,7 @@ Real time interactive streaming digital human， realize audio video synchronous
|
|
|
|
|
|
|
|
|
|
|
|
## News
|
|
|
|
## News
|
|
|
|
- 2024.12.8 完善多并发，显存不随并发数增加
|
|
|
|
- 2024.12.8 完善多并发，显存不随并发数增加
|
|
|
|
|
|
|
|
- 2024.12.21 添加wav2lip、musetalk模型预热，解决第一次推理卡顿问题
|
|
|
|
|
|
|
|
|
|
|
|
## Features
|
|
|
|
## Features
|
|
|
|
1. 支持多种数字人模型: ernerf、musetalk、wav2lip
|
|
|
|
1. 支持多种数字人模型: ernerf、musetalk、wav2lip
|
|
|
|
@ -96,7 +97,6 @@ docker run --gpus all -it --network=host --rm registry.cn-beijing.aliyuncs.com/c
|
|
|
|
- [x] 数字人静音时用一段视频代替
|
|
|
|
- [x] 数字人静音时用一段视频代替
|
|
|
|
- [x] MuseTalk
|
|
|
|
- [x] MuseTalk
|
|
|
|
- [x] Wav2Lip
|
|
|
|
- [x] Wav2Lip
|
|
|
|
- [ ] TalkingGaussian
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
---
|
|
|
|
---
|
|
|
|
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.
|
|
|
|
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.
|
|
|
|
|
lipku
7 months ago