Nguyễn Lê Phúc Vinh
Nguyễn Lê Phúc Vinh
Can you send me the link of the new large-v3 :Đ
> > Can you send me the link of the new large-v3 :Đ > > of course，here，but when I try，tensor error，so I hope you can finish the issue https://openaipublic.azureedge.net/main/whisper/models/e5b1a55b89c1367dacf97e3e19bfd829a01529dbfdeefa8caeb59b3f1b81dadb/large-v3.pt I...
> > > > > 你能把新的 large-v3 的链接发给我吗：Đ > > > > > > > > > > > > 当然，在这里，但是当我尝试时，张量错误，所以我希望你能 https://openaipublic.azureedge.net/main/whisper/models/e5b1a55b89c1367dacf97e3e19bfd829a01529dbfdeefa8caeb59b3f1b81dadb/large-v3.pt 完成这个问题 > > > > > > >...
Can you tell me more details? I still can't figure it out. If you want to talk about the wrong voice energy, I think you should review your dataset, because...
I think your speakers should be recorded with many different pitches, not just 1 pitch, I think this can be fixed
why you choose FFN 3.5x instead of 3x?
Hugging face's tokenizers not support for all non-english language
> I too am convinced that HF already supports training a BPE tokenizer, but I am relatively new to this, could you elaborate? I thought that any text can be...
> yxlllc yxlllc, I've never really appreciated projects developed by the community instead of big companies like OpenAI,... but you've really changed my narrow vision. by what you do. yxlllc,...