Linly-Talker
Linly-Talker copied to clipboard
Digital Avatar Conversational System - Linly-Talker. 😄✨ Linly-Talker is an intelligent AI system that combines large language models (LLMs) with visual models to create a novel human-AI interaction...
感谢作者开源代码,我试用了下musetalk的,好像并没有实现实时对话的效果,流式体现在哪里
# 注意以下几个模块,若安装不成功,可以进入路径利用pip install . 或者 python setup.py install编译安装 # NeRF/freqencoder # NeRF/gridencoder # NeRF/raymarching # NeRF/shencoder 使用上面作者的编译方式,我使用windows系统编译NeRF失败,导致运行后界面里面生成数字人视频失败,请问作者我该如何处理,怎么解决windows系统编译NeRF失败的问题
为什么有些人选择使用纸质地图或寻求方向,而不是依赖GPS设备或智能手机应用程序? 'edge-tts' is not recognized as an internal or external command, operable program or batch file. 函数 TTS_response 运行时间: 1.4796900749206543 秒 函数 LLM_response 运行时间: 1.4827227592468262 秒 using safetensor as default...
webui.py的GPT-SoVITS模型注释掉了默认模型,改成下拉框选择模型,脚本改动如下: ```python # gpt_path = gr.FileExplorer(root = GPT_SoVITS_ckpt, glob = "*.ckpt", value = "s1bert25hz-2kh-longer-epoch=68e-step=50232.ckpt", file_count='single', label="GPT模型路径") # sovits_path = gr.FileExplorer(root = GPT_SoVITS_ckpt, glob = "*.pth", value = "s2G488k.pth", file_count='single', label="SoVITS模型路径")...
webui.py脚本修改成使用textbox ```python gpt_path = gr.Textbox(label="GPT模型路径", value="s1bert25hz-2kh-longer-epoch=68e-step=50232.ckpt") sovits_path = gr.Textbox(label="SoVITS模型路径", value="s2G488k.pth") ``` 加载默认模型正常 ```shell 模型加载中... s1bert25hz-2kh-longer-epoch=68e-step=50232.ckpt s2G488k.pth Number of parameter: 77.49M ``` 改为加载通过GPT-SoVITS训练出的模型报错 ```shell 模型加载中... yk-e15.ckpt yk_e8_s224.pth Traceback (most recent call...
框架中是否考虑增加针对绿幕背景结合其他背景图片实现绿幕的背景替换功能,最终生成可替换背景的数字人视频
update some issues like the shields badges.
Code injection could happen via environment variable. In code [here](https://github.com/Kedreamix/Linly-Talker/blob/c7be21b1b83d6ba227aa6f71bcb1874ee709baec/GPT_SoVITS/inference_webui.py#L51), it directly eval the value from environment variable. A malicous local actor could set something like `export is_half='os.system("touch rickroll")'` to...
wav2lip used too much time on lip change, I found a faster one named Easy-Wav2lip Can author add it to your project? Thanks~
registry.cn-beijing.aliyuncs.com/codewithgpu2/kedreamix-linly-talker:zYGdW3yO3D 如何通过docker compose 启动镜像呢?