You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jul 22, 2024. It is now read-only.
Do you have any plans to provide training the custom words and grammars for this github?
Firstly, I didn't know that just uploading the corpora would generate the custom words automatically.
But when I read the api reference, we can train custom words by adding "sounds_like" and "display_as" fields and grammars so it will be very helpful when those training parts are supported on this github.
This github helped me a lot to understand stt faster and better.
Much appreciated :)
The text was updated successfully, but these errors were encountered:
Adding "sounds_like" entries to out-of-vocabulary words is already supported in the UI by clicking on each individual word in the vocabulary list, but I agree with Ted that having the ability to add individual words and grammars would be useful and would expose more of Watson STT's capabilities.
I will mark this help wanted for now, but will try to get to it if or when I have time.
Do you have any plans to provide training the custom words and grammars for this github?
Firstly, I didn't know that just uploading the corpora would generate the custom words automatically.
But when I read the api reference, we can train custom words by adding "sounds_like" and "display_as" fields and grammars so it will be very helpful when those training parts are supported on this github.
This github helped me a lot to understand stt faster and better.
Much appreciated :)
The text was updated successfully, but these errors were encountered: