Yang Chengyue

Results 19 comments of Yang Chengyue

似乎是nltk这个库的文件有问题,尝试卸载重装看看(?

Above this page, "模型切换" means "Change Model", and you have selected the pretrained model rather than the model you trained here.

> Most likely, the same issue here: python gradio_app/app.py RUNNING ON: cuda Loading pipeline components...: 0%| | 0/6 [00:00

不建议使用cpu,12400+32gb内存 双进程 batchsize为20(把内存吃满)时40s/it

6s的音频合成了25s

1.把 GPT-SoVITS\GPT_SoVITS\prepare_datasets 下三个文件里的 os.environ["CUDA_VISIBLE_DEVICES"] = os.environ.get("_CUDA_VISIBLE_DEVICES") 这一句注释掉 2. 把"GPT-SoVITS\GPT_SoVITS\s2_train.py" 里的"""Assume Single Node Multi GPUs Training Only"""下面一行注释掉 3. 还是上面那个文件,把所有to("mps")改成to("cpu")

抱歉,上一条有缺漏。 s2_train里的os.environ["CUDA_VISIBLE_DEVICES"] = hps.train.gpu_numbers.replace("-", ",") 这一句也要注释掉 在s2_train的main里要手动设置n_gpu以指定开几个进程训练。 s1_train的main里trainer的初始化把accelerator改成cpu,把devices改成1,如果运行gpt训练时出现类型不匹配的问题再把precision改成32

> > 1.把 GPT-SoVITS\GPT_SoVITS\prepare_datasets 下三个文件里的 os.environ["CUDA_VISIBLE_DEVICES"] = os.environ.get("_CUDA_VISIBLE_DEVICES") 这一句注释掉 2. 把"GPT-SoVITS\GPT_SoVITS\s2_train.py" 里的"""Assume Single Node Multi GPUs Training Only"""下面一行注释掉 3. 还是上面那个文件,把所有to("mps")改成to("cpu") > > ![image](https://private-user-images.githubusercontent.com/86220968/301027876-e9d2a12a-3194-4333-93a6-678f358198c6.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.dRa_FTnzJGUBctbGgVN4KX2hx1Ad242EDX9mmzG0m0M) 痾 你好像找错文件夹了,我说的GPT-SoVITS\GPT_SoVITS\prepare_datasets中第一个GPT-SoVITS是有webui.py的那个文件夹

> ![image](https://private-user-images.githubusercontent.com/86220968/301030023-89f2b894-90de-46f7-b7de-d661a5eec408.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.COJexodK5CUbT96dZHBxIesnSAuG_NK_Lh_qjt1Wf7I) 還是一樣 对的,还是这样显示,但是不用管,继续进行后续步骤