Is there any plans in near future to upgrade existing implementation to implement Deepspeech 2.
At this moment there are no firm plans. However, this may change as we consider support of streaming.
What to choose DeepSpeech or DeepSpeech2 , when trying to develop it for some other language and having computation power issue (less no of GPU available) ?
I used DeepSpeech for Urdu language, now, I want to experience DeepSpeech2 ?
Suggestions please.
Thank you!
You’re free to choose whichever variant you wish. However, the repo talked
about in this form uses the DeepSpeech variant from the repo.