flashlight / wav2letter

Facebook AI Research's Automatic Speech Recognition Toolkit
https://github.com/facebookresearch/wav2letter/wiki
Other
6.37k stars 1.01k forks source link

How to get timestamp of each word ? #851

Open manojmsrit opened 4 years ago

manojmsrit commented 4 years ago

I would like to know is there any possible way we can get the timestamp of each word using wave2letter architecture? If so, how should we do it, please let me know regarding the same

tlikhomanenko commented 4 years ago

Please have a look at this tool https://github.com/facebookresearch/wav2letter/tree/v0.2/tools#voice-activity-detection-with-ctc--an-n-gram-language-model

abhinavkulkarni commented 3 years ago

Also, @manojmsrit, take a look at WordUnit class. It is being used inside this utility function to print words, you can modify it to print timestamps as well.

abhinavkulkarni commented 3 years ago

@manojmsrit: You may also want to read the discussion here.