-
Notifications
You must be signed in to change notification settings - Fork 16
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add ctranslate2 whisper models #10
Add ctranslate2 whisper models #10
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Once you make the changes, I can merge and run the evals 🍾
compute_type="float16", | ||
) | ||
|
||
for i in range(3): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What's the outer loop for?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't know. I just copied and pasted this part from the transformers
folder example. 🤗
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ah! Okay, I'll sync and fix this offline.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The outer loop is to compute the average of 3 runs for RTF
…pen_asr_leaderboard into add-ctranslate2-whisper-models
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Dope! Thanks @chainyo
compute_type="float16", | ||
) | ||
|
||
for i in range(3): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ah! Okay, I'll sync and fix this offline.
This PR:
ctranslate2
backend with all the whisper modelsuse_auth_token
totoken
to avoid the warningcalc_rtf.py
I put the
DEVICE_INDEX
as a variable to quickly change it in the shell file forctranslate2
because I had only the device 0 available.Also, the batch feature is currently in beta testing for
faster-whisper
, so the batch is set to 1, but it could be improved soon.I tested the
run_eval.py
script on my side:Fixes #9