Replies: 2 comments
-
|
seems like you are using CPU, I suggest using a smaller model such as https://huggingface.co/facebook/wav2vec2-base-960h |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
Good idea.. Yes I'm using CPU
…On Fri, Jan 10, 2025, 4:35 AM Mahmoud Ashraf ***@***.***> wrote:
seems like you are using CPU, I suggest using a smaller model such as
https://huggingface.co/facebook/wav2vec2-base-960h
—
Reply to this email directly, view it on GitHub
<#52 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AA5GQNRW2BKIKI67XIRN3KL2J6H7FAVCNFSM6AAAAABU4QSOPWVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCNZZGU4TOMQ>
.
You are receiving this because you authored the thread.Message ID:
<MahmoudAshraf97/ctc-forced-aligner/repo-discussions/52/comments/11795972@
github.com>
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I have an english forcedalignemnt task. The quality of results in the repo are amazing, but it takes over 130s to generate_emissions on 36s of audio. A replicate endpoint would be super handy
Beta Was this translation helpful? Give feedback.
All reactions