lipku/metahuman-stream issues and pull requests

#270 - 不同图像尺寸对rtmp推流速度的影响

Issue - State: open - Opened by lokvke 2 months ago

#269 - 怎么接入llm

Issue - State: open - Opened by JiaoYuJiaoYu 2 months ago - 2 comments

#268 - 新需求：怎样使用全篇开口说话的视频来训练数字人？

Issue - State: open - Opened by ThornbirdZhang 2 months ago

#267 - 如何使用https协议来访问https://xxx:8010/rtcpushapi.html,或者使用域名来访问https://service.xxx:8010/rtcpushapi.html

Issue - State: open - Opened by XM233333333 2 months ago - 3 comments

#266 - musev

Issue - State: open - Opened by KelvinHuang66 2 months ago

#265 - musev

Issue - State: closed - Opened by KelvinHuang66 2 months ago

#264 - 我参考作者的实现方式自己实现了GaussianTalker接入进来，为什么asyncio.Queue会一直阻塞住呢？各位懂异步编程的可以帮帮我吗

Issue - State: open - Opened by shehuiwojiege 2 months ago - 3 comments

#263 - 关闭数字人浏览器页面后再打开，显示failed to fetch 后台显示reach max session。

Issue - State: open - Opened by dizhenx 2 months ago - 1 comment

#260 - 大佬，tts我是用的是gpt-sovits。数字人说话时有一顿一顿的杂音，类似电流麦的声音？直接接口调用gpt-sovits的接口没有这种情况发生。

Issue - State: open - Opened by Proletariat-wh 2 months ago - 3 comments

#258 - 大佬们，我采用gpt-vits测试的速度比edgetts要慢很多，人物都无法连续说话了

Issue - State: open - Opened by ThornbirdZhang 2 months ago - 3 comments

#256 - 大模型是流式输出的情况下，tts也用gptsovits流式输出，只会播放最后一句，怎么办

Issue - State: open - Opened by dizhenx 3 months ago - 4 comments

#254 - 接入LLM时报错

Issue - State: open - Opened by zjz0804 3 months ago - 2 comments

#227 - 启动报错

Issue - State: open - Opened by wx24736 3 months ago - 1 comment

#222 - 基于LLM模型实现与metahuman克隆数字人对话，过程总结

Issue - State: open - Opened by wxzz 3 months ago - 5 comments
Labels: good first issue

#147 - 请问运行python app.py --model musetalk --transport webrtc 后一直是start infenrence正常吗

Issue - State: open - Opened by summerprince1 5 months ago - 5 comments

#118 - windows musetalk webrtc问题

Issue - State: open - Opened by TigerHH6866 5 months ago - 5 comments

#106 - AutoDL部署，视频无法加载报错

Issue - State: closed - Opened by jakciehoo 6 months ago - 1 comment

#101 - 浏览器打不开

Issue - State: open - Opened by sukilexi 6 months ago - 5 comments

#100 - 安装pytorch3d报错

Issue - State: open - Opened by Kevin746p 6 months ago - 2 comments

#99 - how to use LLM talking?

Issue - State: open - Opened by TigerHH6866 6 months ago

#98 - Failed to build raymarching-face

Issue - State: open - Opened by sukilexi 6 months ago - 2 comments

#97 - 运行正常了，只有奥巴马的模型？还有其他的模型可以测试吗？

Issue - State: open - Opened by jaysonbai 6 months ago - 5 comments

#96 - 无法合成口型/播放音频

Issue - State: open - Opened by willhzh 6 months ago - 1 comment

#95 - mac M1芯片如何支持？

Issue - State: open - Opened by manwangbin 6 months ago - 2 comments

#94 - 大佬，成功运行后看不到视频

Issue - State: open - Opened by TigerHH6866 6 months ago - 2 comments

#93 - musetalk可以支持rtmp推送吗

Issue - State: open - Opened by during-code 6 months ago - 1 comment

#92 - Feat : 增加 autodl 使用教程

Pull Request - State: closed - Opened by Degree-21 6 months ago - 6 comments

#91 - Fix GPT-SoVITS tts doc

Pull Request - State: closed - Opened by Degree-21 6 months ago - 1 comment

#90 - 关于并发连接的疑问

Issue - State: open - Opened by justiceeem 6 months ago

#89 - 请问可以加字幕吗？即数字人在说话时添加字幕

Issue - State: open - Opened by lemwang 6 months ago - 1 comment

#88 - rtcpush使用问题

Issue - State: open - Opened by jackietu 6 months ago

#87 - 请教websocket 外网不能连接问题

Issue - State: open - Opened by wingjoezhou 6 months ago - 3 comments

#86 - docker运行报错

Issue - State: open - Opened by FLH1030 6 months ago - 5 comments

#85 - 遇到跨域的问题：No 'Access-Control-Allow-Origin' header

Issue - State: open - Opened by aiquanpeng 6 months ago - 1 comment

#84 - 执行App.py时报错

Issue - State: open - Opened by LiuJiaqi0505 6 months ago - 4 comments

#83 - ernerf ----> synctalk Thanks.

Issue - State: open - Opened by jeffreyhaoau 6 months ago - 1 comment

#82 - 运行python app.py报错

Issue - State: closed - Opened by mkygogo 7 months ago - 5 comments

#81 - Win10, Docker安装报错

Issue - State: open - Opened by mkygogo 7 months ago - 3 comments

#80 - werkzeug.routing.exceptions.WebsocketMismatch: 400 Bad Request: The browser (or proxy) sent a request that this server could not understand.

Issue - State: open - Opened by weibiaoyi 7 months ago - 4 comments

#78 - start websocket server Error: Cannot connect to host localhost:1985 ssl:default [Connect call failed ('127.0.0.1', 1985)]

Issue - State: open - Opened by Akatukiaoki 7 months ago - 4 comments

#77 - webRtc使用问题

Issue - State: closed - Opened by Pilot-Labs-Dev 7 months ago - 4 comments

#76 - 在替换自己模型的时候报参数不一致错误，应该怎么修改或者去哪里修改

Issue - State: closed - Opened by zangruoxiang1 7 months ago - 8 comments

#75 - 声音卡顿问题

Issue - State: open - Opened by hqq93 7 months ago - 3 comments

#74 - 声音断续问题

Issue - State: open - Opened by angelbi 7 months ago - 3 comments

#73 - 使用gpt-sovits。报错

Issue - State: open - Opened by hqq93 7 months ago - 2 comments

#72 - “docker版本已经不是最新代码，可以作为一个空环境，把最新代码拷进去运行”是把代码拷贝到docker容器里面的“/nerfstream”文件夹里面吗

Issue - State: open - Opened by gavid0124 7 months ago - 2 comments

#71 - 执行pip install tensorflow-gpu==2.8.0时报错了

Issue - State: open - Opened by gavid0124 7 months ago - 1 comment

#70 - OMP: Error #13: Assertion failure at kmp_affinity.cpp(4314).

Issue - State: open - Opened by kingsaction 7 months ago

#69 - Webrtc 视频显示不出来

Issue - State: open - Opened by zkxd2018 7 months ago - 9 comments

#68 - Error in step python app.py

Issue - State: open - Opened by h-oneohone 7 months ago - 2 comments

#67 - 提交全身图片动态缓存池写法

Pull Request - State: closed - Opened by yanyuxiyangzk 7 months ago - 1 comment

#66 - 使用docker，rtmpserver (srs)和app.py都运行起来，但vlc没有输出

Issue - State: open - Opened by s3219521aa 7 months ago - 1 comment

#65 - fatal error C1189: #error: You need C++17 to compile PyTorch nvcc warning : incompatible redefinition for option 'std', the last value of this option was used freqencoder.cu ninja: build stopped: subcommand failed.

Issue - State: open - Opened by mbt1909432 7 months ago - 2 comments

#64 - GPU是必须的吗

Issue - State: open - Opened by wqw547243068 7 months ago - 1 comment

#63 - webrtc得到的视频声音有很明显的电流声

Issue - State: closed - Opened by zhangdanfeng888 7 months ago - 3 comments

#62 - webrtc

Issue - State: closed - Opened by huyppppppp 7 months ago - 3 comments

#61 - 更新到最新版本后，python app.py可以正常运行，但是在网页端点击send后报错

Issue - State: closed - Opened by zhangdanfeng888 7 months ago - 1 comment

#60 - 修复关于qwen的一些代码错误

Pull Request - State: closed - Opened by kuun993 8 months ago

#59 - 修正使用webrtc推流时全身问题

Pull Request - State: closed - Opened by omanhom 8 months ago

#58 - 效果很差声音卡顿

Issue - State: closed - Opened by Hujiazeng 8 months ago - 1 comment

#57 - 升级基于gpt_sovits的接口和语言识别demo演示

Pull Request - State: closed - Opened by yanyuxiyangzk 8 months ago

#56 - 糟糕透了

Issue - State: closed - Opened by Hujiazeng 8 months ago

#55 - 人物衣服颜色变化

Issue - State: open - Opened by huyppppppp 8 months ago - 3 comments

#54 - 项目优化，大模型推理加速，可以提高本地大模型响应速度，同时优化了部署环境，用全新docker环境部署不容易出错

Pull Request - State: closed - Opened by yanyuxiyangzk 8 months ago

#53 - [aac @ 0x7f169587a4c0] Specified sample format -1 is invalid or not supported Could not open audio encoder!请问一下这个报错应该如何解决

Issue - State: open - Opened by xdujinsiyu 8 months ago - 1 comment

#52 - 请问支持绿幕背景吗？

Issue - State: open - Opened by huangjianyi0701 8 months ago - 1 comment

#51 - nerfreal.py中的warm_up()是必要的嗎

Issue - State: open - Opened by YoranLiu 8 months ago - 2 comments

#50 - SRS镜像

Issue - State: closed - Opened by huyppppppp 8 months ago - 3 comments

#49 - 全身视频贴图后，切割线处对应不上

Issue - State: closed - Opened by redstarxz 8 months ago - 5 comments

#48 - 运行python app.py报错

Issue - State: closed - Opened by zhangdanfeng888 8 months ago - 6 comments

#47 - win10+Anaconda环境下如何搭建

Issue - State: open - Opened by hqq93 8 months ago - 4 comments

#46 - 大佬，请问推流怎么加上身子呢？

Issue - State: open - Opened by zhuxiu1234 9 months ago - 2 comments

#45 - 替换成自己训练的模型报错

Issue - State: open - Opened by feipengheart 9 months ago - 4 comments

#44 - GPU RAM最低要求是多少，6G可用吗

Issue - State: open - Opened by ivoidcat 9 months ago - 2 comments

#43 - Cant get any video stream on Vlc with rtmp://localhost/live/livestream, err with Requested output format 'flv' is not a suitable output format

Issue - State: closed - Opened by nypyp 9 months ago - 1 comment
Labels: good first issue

#42 - vlc拉流，播放延迟的问题

Issue - State: open - Opened by mrlihellohorld 9 months ago - 1 comment

#41 - installation steps ( 安装步骤)

Issue - State: open - Opened by shailesh-growexxer 9 months ago - 6 comments

#40 - 请问可以将背景透明吗？

Issue - State: open - Opened by zhangchenggogo 9 months ago - 3 comments

#39 - 请问声音为什么一顿一顿的

Issue - State: open - Opened by azraeljian 9 months ago - 16 comments

#38 - 执行python app.py报错

Issue - State: open - Opened by OpenSorceYCW 9 months ago - 3 comments

#37 - pyaudio调用本地麦克风流式推理能实现吗？

Issue - State: open - Opened by ThetaRgo 9 months ago

#36 - 求助运行python app.py出现如下错误

Issue - State: closed - Opened by azraeljian 9 months ago - 2 comments

#35 - 推流延迟问题

Issue - State: open - Opened by lswc225 9 months ago - 6 comments

#34 - OSError: CUDA_HOME environment variable is not set

Issue - State: closed - Opened by azraeljian 9 months ago - 1 comment

#33 - OSError: CUDA_HOME environment variable is not set

Issue - State: closed - Opened by azraeljian 9 months ago

#32 - 修改python/site-packages/werkzeug/routing/matcher.py，注释websocket_mismatch = True

Issue - State: open - Opened by OpenSorceYCW 9 months ago - 7 comments

#31 - 本地网页端输入内容，生成数字人区域无反应？

Issue - State: open - Opened by OpenSorceYCW 9 months ago - 7 comments

#30 - 效果怎么样？有训练代码吗？

Issue - State: open - Opened by dizhenx 9 months ago

#29 - 推流后如何添加背景图

Issue - State: open - Opened by BeanBunWang 9 months ago - 3 comments

#28 - 推流语速问题

Issue - State: closed - Opened by lswc225 9 months ago - 5 comments

#27 - 输入文本后数字人没有变化

Issue - State: closed - Opened by xren0085 9 months ago - 1 comment

#26 - 请问生成的音频数据，如何保存为wav文件呢

Issue - State: open - Opened by mrlihellohorld 9 months ago - 1 comment

#25 - 推流失败，没有生成视频

Issue - State: closed - Opened by redstarxz 9 months ago

#24 - 推流失败，没有生成视频

Issue - State: closed - Opened by redstarxz 9 months ago - 1 comment

#23 - 博主，一些憨憨拿你的代码在B站上割韭菜

Issue - State: open - Opened by yyheart 10 months ago - 1 comment

#22 - 大佬，为什么模特不眨眼了，是哪个参数需要设置？

Issue - State: open - Opened by stevin-dong 10 months ago - 4 comments

GitHub / lipku/metahuman-stream issues and pull requests