Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / RVC-Boss/GPT-SoVITS issues and pull requests
#2074 - 允许使用共享显存开启V3训练
Pull Request -
State: open - Opened by KamioRinn about 18 hours ago
#2073 - Patch-LangSegmenter
Pull Request -
State: open - Opened by KamioRinn about 19 hours ago
#2072 - V3点了很多次无法更换GPT和SOVITS模型
Issue -
State: closed - Opened by AkebiKomichi about 20 hours ago
#2071 - 后端调用api时出现307状态码,问问佬们怎么解决
Issue -
State: open - Opened by skyspring18 about 22 hours ago
- 1 comment
#2070 - 报错,1C推理打不开。
Issue -
State: open - Opened by AkebiKomichi about 22 hours ago
- 4 comments
#2069 - 发现关于langsegmenter.py的一个小bug
Issue -
State: open - Opened by Rafa00127 about 24 hours ago
#2068 - Update requirements.txt
Pull Request -
State: open - Opened by Lion-Wu 1 day ago
- 1 comment
#2067 - CPU 使用更低 batch size
Pull Request -
State: closed - Opened by Lion-Wu 1 day ago
#2066 - 佬,是否有计划新增api_v3?
Issue -
State: open - Opened by ltbjwzttzz 1 day ago
#2065 - v3 为什么采样步数调到32都感觉音质不太好,有沙沙的声音
Issue -
State: open - Opened by wenHK 1 day ago
- 1 comment
#2064 - Python API?
Issue -
State: open - Opened by ZhaoyuanQiu 1 day ago
#2063 - 打开推理UI报错,如何解决
Issue -
State: open - Opened by benniechenbin 1 day ago
- 1 comment
#2062 - Add en_normalization and fix LangSegmenter
Pull Request -
State: closed - Opened by KamioRinn 2 days ago
#2061 - fix inference issue
Pull Request -
State: closed - Opened by wzy3650 2 days ago
#2060 - 你们好像没有更新TTS_Infer_pack 导致API v2 用不了v3模型 , 我给inference_webui 打包了个 兼容api.
Issue -
State: open - Opened by Kas1o 2 days ago
#2059 - 关于梯度检查点会不会对模型劣化的问题
Issue -
State: open - Opened by KakaruHayate 2 days ago
- 4 comments
#2058 - [V3 BUG]ZeroDivisionError: integer modulo by zero
Issue -
State: open - Opened by Pevernow 2 days ago
- 3 comments
Labels: In follow-up
#2057 - 关于V3和V2的一些问题
Issue -
State: open - Opened by AIBuilder2023 3 days ago
- 1 comment
#2056 - Phoneme duration tracking
Issue -
State: closed - Opened by davidchi31415 3 days ago
- 1 comment
#2055 - 运行api_v2.py,post文本语言为zh(中英混合),会INFO: 127.0.0.1:53391 - "POST /tts HTTP/1.1" 400 Bad Request
Issue -
State: open - Opened by jianerwuling 3 days ago
#2054 - 启动api后声音模型设置自动回退到默认值
Issue -
State: open - Opened by wxhou 4 days ago
- 1 comment
#2053 - v3 VS v2
Issue -
State: open - Opened by foreverhell 4 days ago
- 3 comments
#2052 - I couldn't find API access to v3 version?
Issue -
State: open - Opened by DurmaTrading 4 days ago
- 2 comments
#2051 - Regarding the API of v3
Issue -
State: open - Opened by eymenongelsama 4 days ago
- 1 comment
#2050 - 当模型默认为v3的时候会提示缺少模型
Issue -
State: closed - Opened by aiguoliuguo 4 days ago
- 7 comments
#2049 - 修改默认模型选择为v3,并且规范了模型版本的判断逻辑
Pull Request -
State: closed - Opened by aiguoliuguo 4 days ago
- 1 comment
#2048 - Missing keys when loading v3 model checkpoint for semantics token extraction
Issue -
State: closed - Opened by justinjohn0306 4 days ago
- 2 comments
#2047 - 切换新的语言分割工具
Pull Request -
State: closed - Opened by KamioRinn 5 days ago
- 1 comment
#2046 - Some error while trying to get SoVITS v3 running on Mac
Issue -
State: open - Opened by leleayaka 5 days ago
- 5 comments
#2045 - 低端机无法利用CUDA torch.float16推理v3 Low-end machines unable to inference v3 using CUDA torch.float16
Issue -
State: open - Opened by Sewlell 5 days ago
- 1 comment
#2044 - 关于频谱
Issue -
State: closed - Opened by ajlsunset 5 days ago
- 2 comments
#2043 - Update inference_webui.py, 修复在Colab使用自行训练的v2模型时可能遇到的UnpicklingError: Weights only load failed.
Pull Request -
State: closed - Opened by Abyss-Seeker 5 days ago
- 1 comment
#2042 - V3的整合包什么时候会出?
Issue -
State: closed - Opened by jackwong25 5 days ago
- 5 comments
#2041 - About V3
Issue -
State: open - Opened by leleayaka 6 days ago
- 1 comment
#2040 - update Gradient Checkpointing to reduce VRAM usage
Pull Request -
State: closed - Opened by KakaruHayate 6 days ago
#2039 - 在推理页面选SoVITS模型列表,提示出错。
Issue -
State: open - Opened by fuqingbin 6 days ago
- 12 comments
Labels: In follow-up
#2038 - V3的模型为什么会到750M这么大?效果相对于V2提升很多吗?
Issue -
State: open - Opened by AnonymousmousCoder 6 days ago
- 3 comments
Labels: In follow-up
#2037 - 'hps' is not defined Error
Issue -
State: closed - Opened by sjxm0721 6 days ago
- 5 comments
#2036 - V3 - Missing model file s2Dv3.pth
Issue -
State: closed - Opened by dignome 6 days ago
- 1 comment
#2035 - V 3 版本,有api吗
Issue -
State: open - Opened by fxyyds 6 days ago
#2034 - remove unnecessary mask of batch infer
Pull Request -
State: open - Opened by wzy3650 6 days ago
- 2 comments
#2033 - Improve Japanese translation
Pull Request -
State: closed - Opened by Fyphen1223 6 days ago
#2032 - Update Documentation
Pull Request -
State: closed - Opened by SapphireLab 7 days ago
#2031 - windows 系统,torchmetrics 和 matplotlib 环境冲突问题
Issue -
State: closed - Opened by flyskywalkerlby 7 days ago
#2030 - 使用打包文件出现“ImportError: cannot import name 'deprecated' from 'typing_extensions'”
Issue -
State: open - Opened by PerviousFurther 7 days ago
#2029 - GPT训练时出现“RuntimeError: Index tensor must have the same number of dimensions as self tensor”
Issue -
State: open - Opened by PerviousFurther 7 days ago
- 2 comments
#2028 - 使用 api_v2.py 运行 后 音频没有声音,但用webui 启动一切正常
Issue -
State: open - Opened by ModuleCode 8 days ago
#2027 - 我想在已经训练出的模型上用新加的语音继续训练,应该在哪里修改些什么呢
Issue -
State: open - Opened by MrParzifal 8 days ago
- 2 comments
#2026 - 在搭载RTX 50系显卡的Windows系统上部署GPT-SoVITS(WSL2方案)
Issue -
State: closed - Opened by syd2333 8 days ago
- 2 comments
#2025 - 使用字幕标注,校准时无法读取音频
Issue -
State: open - Opened by PerviousFurther 8 days ago
#2024 - 自己录制的声音,不能很好的克隆播放,经常把模板的里面重复读
Issue -
State: open - Opened by hjj-lmx 8 days ago
- 1 comment
#2023 - 关于整理打标时批量删除未识别出的音频的功能的问题。
Issue -
State: closed - Opened by NaughtDZ 8 days ago
- 1 comment
#2022 - 无法获取配置文件,请问如何解决?
Issue -
State: open - Opened by 1755508167 8 days ago
- 1 comment
#2021 - 可否添加语音转语音的功能?或者训练好的模型是否可以对接RVC project进行音色复制?
Issue -
State: open - Opened by Ctree35 9 days ago
#2020 - 标题:GPT-SOVITS-V2在某些英文单词的发音上存在问题 | Issues with Pronunciation of Certain English Words in GPT-SOVITS-V2
Issue -
State: open - Opened by S-T-K 10 days ago
- 3 comments
#2019 - 微调时有两行报错,cuda占用为0。
Issue -
State: closed - Opened by yudonw 10 days ago
- 1 comment
#2018 - api可以返回带时间戳的音素数据吗
Issue -
State: open - Opened by tangclQaQ 10 days ago
- 1 comment
#2017 - 多人物对话能否实现
Issue -
State: open - Opened by rockManFlow 10 days ago
- 1 comment
#2016 - MAC 无法打开 TTS 推理的 WebUI
Issue -
State: open - Opened by lemonsz15 10 days ago
- 2 comments
#2015 - 如果文字的数量不多,生成很不稳定,有时候是空语音,有时候返回的是模板语音里面的,这种怎么解决
Issue -
State: open - Opened by hjj-lmx 10 days ago
- 2 comments
#2014 - Is it possible to add other languages?
Issue -
State: closed - Opened by marvinbelfort 11 days ago
- 1 comment
#2013 - api.py 没法生成音频,api_v2.py 也没法启动
Issue -
State: open - Opened by Wu1905 11 days ago
- 1 comment
#2012 - 纯英文的读起来很奇怪,是不是因为 bert 全为 0 导致的?能否优化?
Issue -
State: open - Opened by L-jasmine 13 days ago
#2011 - 请问有办法直接支持拼音吗
Issue -
State: closed - Opened by jasonzhang761213 13 days ago
- 3 comments
#2010 - optimize attention calc logic
Pull Request -
State: closed - Opened by wzy3650 13 days ago
- 1 comment
#2009 - error Connection errored out.
Issue -
State: closed - Opened by iwundu-hmy 13 days ago
- 1 comment
#2008 - 【求助诸位大佬】推理时总是出现合成音频为空音频或合成的音频中断
Issue -
State: open - Opened by Simon568667 13 days ago
#2008 - 【求助诸位大佬】推理时总是出现合成音频为空音频或合成的音频中断
Issue -
State: open - Opened by Simon568667 13 days ago
#2007 - About Integration of visual features in GPT-SoVITS pipeline
Issue -
State: open - Opened by Lokesh-Kumar1998 13 days ago
#2007 - About Integration of visual features in GPT-SoVITS pipeline
Issue -
State: open - Opened by Lokesh-Kumar1998 13 days ago
#2006 - 4090D这个推理耗时正常吗
Issue -
State: open - Opened by TheoTian 13 days ago
- 2 comments
Labels: In follow-up
#2005 - 无法import opencc
Issue -
State: closed - Opened by rainbowcandy 13 days ago
- 2 comments
#2004 - cannot recognise 3060GPU while using UVR5
Issue -
State: open - Opened by belululuga 16 days ago
- 1 comment
Labels: In follow-up
#2003 - torch v2.6.0不适配代码,应更新requirements
Issue -
State: closed - Opened by Xraytest 17 days ago
- 6 comments
#2002 - colab上GPT训练错误
Issue -
State: closed - Opened by Ebotian 23 days ago
- 2 comments
Labels: In follow-up
#2001 - 建议制作一个一键托管至koyeb的按钮
Issue -
State: open - Opened by HRJJACK1145 23 days ago
#2000 - Compatible with older graphics cards
Pull Request -
State: closed - Opened by v3ucn 23 days ago
#1999 - 在 api_v2 中使用韩语请求时,文本在前端处理过程中消失的错误
Issue -
State: open - Opened by yellow0519 24 days ago
#1998 - asr遇到问题
Issue -
State: open - Opened by zlinwzx147258 25 days ago
- 1 comment
#1997 - 无法定位程序输入点
Issue -
State: open - Opened by LaciliaExe 25 days ago
#1996 - 缺少文件,但文件存在
Issue -
State: open - Opened by ZhangQL2824 25 days ago
- 1 comment
#1995 - 1C的两个模型列表无法展开
Issue -
State: open - Opened by awa-qwq-TAT2022 25 days ago
- 2 comments
#1994 - I am stuck in 1Aa-Text training on Colab.
Issue -
State: open - Opened by hgzs233 26 days ago
#1993 - 无法开启TTS推理WebUI
Issue -
State: open - Opened by PatrickQia 27 days ago
#1992 - TTS Results Occasionally Truncated
Issue -
State: open - Opened by KangSquad 27 days ago
#1991 - AttributeError: 'FigureCanvasAgg' object has no attribute 'tostring_rgb'
Issue -
State: closed - Opened by hjcxlp 27 days ago
- 2 comments
#1990 - 关于api的使用和版本问题
Issue -
State: closed - Opened by TheShy-Dream 28 days ago
- 1 comment
#1989 - 返回的音频出现非常长时间的噪音和重复
Issue -
State: open - Opened by XiHenSuper 28 days ago
#1988 - 发现对于mac其实是在用cpu推理,在改成mps后反而速度降了,cpu是大概七十it,但是mps只有二十多,这是为什么?
Issue -
State: open - Opened by 227uijk 29 days ago
- 1 comment
#1987 - archlinux采用脚本: bash install.sh 方式安装报错
Issue -
State: closed - Opened by shaoooping 29 days ago
- 1 comment
#1986 - what is the simple step to add a new language support?
Issue -
State: open - Opened by dream-1ab 30 days ago
#1985 - 【求助各位大佬】Colab训练GPT模型时报错
Issue -
State: closed - Opened by hsgwktb about 1 month ago
- 11 comments
Labels: In follow-up, todolist
#1984 - linux conda env 无法进行GPT训练
Issue -
State: closed - Opened by MintGrass about 1 month ago
- 2 comments
#1983 - 更好的 G2p
Issue -
State: open - Opened by L-jasmine about 1 month ago
#1982 - RuntimeError: use_libuv was requested but PyTorch was build without libuv support
Issue -
State: open - Opened by clodiamery about 1 month ago
#1981 - 官方权重 及 更多数据继续训练
Issue -
State: closed - Opened by Bobchenyx about 1 month ago
- 1 comment
#1980 - 使用调用GPT_SoVITS API 克隆自己的声音时没有返回target语料的音频,但是会返回参考语料对应的音频。
Issue -
State: closed - Opened by huazai-1994 about 1 month ago
#1979 - Everton
Issue -
State: closed - Opened by kidgalerhad about 1 month ago
#1937 - Question about 2-get-hubert-wav32k
Issue -
State: closed - Opened by Shijun-droid about 1 month ago
- 3 comments
#1936 - 连续多次请求串音色问题
Issue -
State: open - Opened by DBDXSS about 1 month ago
- 1 comment