hz

For simplicity, this PR is merged with a minimal implementation in #303

1

For simplicity, this PR is merged with a minimal implementation in #303 _Originally posted by @hwidong-na in https://github.com/speechbrain/speechbrain/issues/288#issuecomment-695081709_

loss nan

1

训练过程中，极易出现loss nan

### 是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this? - [X] 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions ### 该问题是否在FAQ中有解答？ | Is there an...

想添加noise token训练模型，在config.yaml里添加了添加了  并设置ignore_init_mismatch: true 添加noise token之后训练，请问text文件应该是什么样的？ 

create wss with C , find some more info in the end

3

### Is there an existing issue for this? - [ ] I have searched the existing issues ### Describe the bug create wss with C , find some more info...

今天上午git clone了最新代码，不管输入文本是什么，tts结果都是一样的，且都听不懂。

[BUG]生成的音频出现拼接问题

1

python tools/llama/generate.py \ --text "海鸥岛有哪些好玩的地方" \ --prompt-text "打开晾衣架照明" \ --prompt-tokens "fake.npy" \ --checkpoint-path "results/text2semantic_400m_finetune_spk/checkpoints/step_000000400.ckpt" \ --speaker "1" \ --num-samples 2 \ --compile 用自己的语料微调了一下vqgan & llama prompt对应的音频是：打开晾衣架照明 生成的音频内容是：哎。。。打开晾衣架照明，海鸥岛有哪些好玩的地方