mozilla / DeepSpeech

DeepSpeech is an open source embedded (offline, on-device) speech-to-text engine which can run in real time on devices ranging from a Raspberry Pi 4 to high power GPU servers.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Word Probabilities

alex-karim-eladl opened this issue · comments

Hi, I'm a quite new to these kinds of models so I'd appreciate your help! I want to rate speech clarity in addition to transcribing the words. I know the RNN converts an input to a sequence of character probabilities. I was wondering if it would be possible to output a probability along with each outputted word from this specific model? If so, how should I go about making these changes? Again, thank you for your tips!

Take a look at issue #3693