Skip to content
Discussion options

You must be logged in to vote

It's better to use a Chinese corpus-pretrained HuBERT instead of the original English version. You might want to try this resource: https://github.com/TencentGameMate/chinese_speech_pretrain.

Regarding BBPE finetuning, I don't have experience with it, but using the same letter-level vocabulary from Aishell-1 works for me."

Replies: 1 comment 3 replies

Comment options

You must be logged in to vote
3 replies
@redsheep913
Comment options

@yfyeung
Comment options

yfyeung May 30, 2025
Collaborator

@redsheep913
Comment options

Answer selected by yfyeung
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants