lzl1456
lzl1456
@nonmetal Hello, have you solved this problem? The keams model I trained for 1000 hours is not working either. The labels from the data are basically the same. Is it...
“”CT-Transformer标点-中英文-通用-large“” 这个CT-tranformer 标点的模型，并不是采用的看未来L长度呀，如果输入text_lengths按每条长度，里面代码就只乘了这个做成的mask，就是看全局的san_M， 有没有跟论文一样的配置代码模型呢？ 还有该模型前向20切分输入，是否是最佳值呢
> 实现了一般，效果不行，应该有问题，等他开源吧
oh Pity ！ I extracted the token in advance,so I don't need to import the hubert model 。 but But still this problem occurs when use accelerate launch multi-GPu train...
I think I solved the problem the class Attention(nn.Module): add code if num_null_kv > 0: self.null_kv = nn.Parameter(torch.randn(2, num_null_kv, dim_head))
> @scf4 Hi, we are using the same fairseq codebase as the speech HuBERT, but we only released the huggingface checkpoint for now. It seems like audiolm-pytorch is using the...