4 Pull requests merged by 2 people

• Remove Debug Code

  #2471 merged Jun 18, 2025

• Fix bugs in install.sh, reduce log noise, and improve error reporting

  #2464 merged Jun 17, 2025

• 优化 torch_script 导出模型

  #2460 merged Jun 13, 2025

• export_torch_script.py support v2Pro & v2ProPlus

  #2456 merged Jun 12, 2025

4 Pull requests opened by 4 people

• feat: implement responsive top banner with dark/light mode support

  #2455 opened Jun 12, 2025

• fix bug

  #2463 opened Jun 14, 2025

• 尝试 stream infer

  #2469 opened Jun 16, 2025

• Fixed syntax errors for api_v2.py

  #2473 opened Jun 18, 2025

5 Issues closed by 3 people

• v2Pro 和 v2ProPuls 使用相同的 S1 (GPT) 模型嗎？

  #2470 closed Jun 17, 2025

• 导出v2pro的torchscript模型的时候，在export_torch_script.py中的dict_s2 = torch.load(vits_path, weights_only=False) 处报错，这个怎么处理啊

  #2438 closed Jun 16, 2025

• 未开发

  #2459 closed Jun 13, 2025

• AttributeError: 'NoneType' object has no attribute 'half'

  #2461 closed Jun 13, 2025

• Please add

  #2457 closed Jun 13, 2025

7 Issues opened by 6 people

• 最新的代码中，能直接支持多GPU推理吗？通过改config的配置文件

  #2472 opened Jun 18, 2025

• 推理特化包中接口文档中POST /infer_single 中的 "app_key": "", "dl_url": ""该填哪些内容？

  #2468 opened Jun 16, 2025

• 关于一些加速做法的询问

  #2467 opened Jun 15, 2025

• 从零训练sovits模型遇到的音质问题

  #2466 opened Jun 15, 2025

• 关于无法使用微调模型的问题

  #2465 opened Jun 15, 2025

• 4080s v2模型 sovits训练速度2it/s

  #2462 opened Jun 13, 2025

• 请问一下有人遇到过长文本，出现显存不释放吗？

  #2458 opened Jun 13, 2025

8 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Broken v1 finetuning

  #2451 commented on Jun 12, 2025 • 0 new comments

• 处理超长文本时出现GPU爆显存问题

  #1519 commented on Jun 12, 2025 • 0 new comments

• api_v2.py的粤语合成尝试batch_size并行推理，但填写了prompt_text后，输出的合成语音是参考音频的重复循环。而非待输入合成的text，请大佬帮忙看下。感谢！

  #2415 commented on Jun 12, 2025 • 0 new comments

• 文字包含化学表达式时怎么处理呀

  #1229 commented on Jun 16, 2025 • 0 new comments

• 中文输入但输出的是日语语音,这个是什么问题

  #2263 commented on Jun 16, 2025 • 0 new comments

• 改造 GPT-SoVITS, 命令行取代web ui, 支持多语言训练

  #2430 commented on Jun 17, 2025 • 0 new comments

• 想要训练AMSR应该如何建立数据集？How should I build a dataset to train AMSR?

  #2382 commented on Jun 17, 2025 • 0 new comments

• WebUI简化方案

  #2399 commented on Jun 18, 2025 • 0 new comments