BQ-Wang0511/TalkLikeYou-Streaming-RealTime-Demo

选用指南

AI 根据 README 整理 · 2026/10/14 17:15:33

使用方式

通过命令行启动服务器，然后在浏览器中打开界面使用；也提供命令行客户端。

查看 README 依据

cd TalkLikeYou-Streaming-RealTime-Demo CUDA_VISIBLE_DEVICES=0 python app.py --host 0.0.0.0 --port 5070 ``` Open `http://127.0.0.1:5070`. Uploading a source starts one-time avatar registration; subsequent runs reuse the cached features. Image sources are used by Audio and Chat agent, while video sources are used by Video dubbing. No motion-reference file is required for a video source. ### Audio

Configure the services below, select **Chat agent**, upload an image avatar, and enter text or record speech. The response text, synthesized speech, and avatar frames are streamed incrementally. Browser speech recognition may be used when available; otherwise the configured STT endpoint handles recorded audio. The equivalent launcher is: ```bash CUDA_VISIBLE_DEVICES=0 ./run.sh ``` The command-line client covers Audio and Video dubbing: ```bash python demo_client.py --source path/to/source.jpg --audio path/to/audio.wav

部署要求

需要 Linux、NVIDIA GPU 和 FFmpeg，测试环境为 Python 3.10、CUDA 12.1 和 cuDNN 9。

查看 README 依据

All three modes share cached avatar preprocessing and the same streaming renderer. Audio and Chat agent accept image avatars; Video dubbing accepts video avatars and does not require a motion-reference or habit-reference file. ## Installation Linux, an NVIDIA GPU, and FFmpeg are required. The tested environment uses Python 3.10, CUDA 12.1, and cuDNN 9. ```bash conda env create -f environment.yaml conda activate talklikeyou-demo ``` Alternatively, install a CUDA-compatible PyTorch build and then run: ```bash pip install -r requirements.txt ``` ## Run ```bash

已知限制

作者指出，将论文模型的 motion_generator.pt 重命名为 lip_motion.pt 会导致错误，因为两者模型定义和参数不同。

查看 README 依据

The default paths can be overridden with `--lip_ckpt`, `--pose_ckpt`, `--data_root`, and `--cfg_pkl`. Do not rename `motion_generator.pt` to `lip_motion.pt`; their model definitions and parameters differ. ## Acknowledgements The portrait renderer incorporates components adapted from [Ditto](https://github.com/antgroup/ditto-talkinghead), and portrait preprocessing incorporates components adapted from [LivePortrait](https://github.com/KwaiVGI/LivePortrait). We thank both teams for releasing their work. See [Third-Party Notices](THIRD_PARTY_NOTICES.md) for the applicable licenses. ## License