seamless_communication icon indicating copy to clipboard operation
seamless_communication copied to clipboard

Published 20 hours ago • verified publisher icon facebookresearch

Source language detection

Open AravAct opened this issue 1 year ago • 4 comments

How to get source language detection. This should be similar to detect_language function in whisper Reason: When using a chatbot I want to automatically detect the source language and provide the final answer in the source language directly.

AravAct avatar Nov 02 '23 09:11 AravAct

@Vaibhavs10 , Is there anyway to do this? thanks!

iamshreeram avatar Dec 03 '23 13:12 iamshreeram

Would also be interested.

asusdisciple avatar Dec 05 '23 11:12 asusdisciple

This is a valuable Whisper feature that is really missing from m4t. The value of the model is much lower if we can't know what was the original language.

Please make this available.

aberaud avatar Dec 31 '23 13:12 aberaud

The Seamless project did not release a speech language identification model. However, you can use a speech LID model from a related project called MMS: https://github.com/facebookresearch/fairseq/blob/main/examples/mms/README.md#lid.

In https://github.com/facebookresearch/seamless_communication/issues/325 I give some more details.

avidale avatar Jan 18 '24 09:01 avidale