hildazzz

Results 22 comments of hildazzz

感谢及时回复,可以正常使用了。冒昧问下两个分支的差别是什么?

我要也要自制多人语料库(非标贝)用来训练,用Bert_vits这个分支可以吗?具体预处理是需要参考vits_prepare.py对吗?

me too! using the pretrained weights (charactr/vocos-mel-24khz), i have tried several audio samples, the same problem occurred.

> > 是否是在相同的prompt下进行对比,cosyvoice和coysvoice2初始化时,都指定load_jit等为false > > 把coysvoice2初始化的load都设置成false之后速度确实变快了,推理“你好,请稍等,我正在思考您的问题…”的时间是3秒。 所以使用TorchScript模型反而没有加速吗?

同问,请问语气控制可以出个说明吗?具体支持种类,以及对应的效果之类的

> > > > 请问有flow模型的训练损失图吗? flow模型的from scratch训练最后损失大概什么水平? > > > > > > > > > flow 损失是震荡下降的 > > > > > > flow模型一共训练了多少个步骤呢?是SFT训练还是不依赖基础模型训练的? > > flow 模型损失是震荡下降,sft 方式训练,epoch...

> Hi, the current MaskGCT supports Chinese (in fact, we support six languages: en, zh, fr, de, kr, ja), can you give me more details about the error, for example,...

> [#729 (comment)](https://github.com/FunAudioLLM/CosyVoice/issues/729#issuecomment-2545399338) > > * [CosyVoice2-0.5B does not have spk2info.pt #729](https://github.com/FunAudioLLM/CosyVoice/issues/729) > > It appears that you're encountering an issue with the absence of the spk2info.pt file in the...

我的torch版本是2.2.1,也会报这个错误。请问如何解决呢?