search

facebookresearch / seamless_communication

Foundational Models for State-of-the-Art Speech and Text Translation
Other
10.81k stars 1.05k forks source link

Source language detection #206

Open AravAct opened 11 months ago

AravAct commented 11 months ago

How to get source language detection. This should be similar to [detect_language function in whisper](https://github.com/openai/whisper/blob/main/README.md#:~:text=Below%20is%20an%20example%20usage%20of%20whisper.detect_language()%20and%20whisper.decode()%20which%20provide%20lower%2Dlevel%20access%20to%20the%20model.) Reason: When using a chatbot I want to automatically detect the source language and provide the final answer in the source language directly.

iamshreeram commented 10 months ago

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

asusdisciple commented 10 months ago

Would also be interested.

aberaud commented 9 months ago

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.

avidale commented 8 months ago

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.