Can this implementation of DeepSpeech 2 be integrated to Mozilla’s DeepSpeech?
Deepspeech is no longer under active development after the Mozilla layoffs. But big parts of the former team founded the company Coqui AI where they forked Deepspeech and keep on developing it: https://github.com/coqui-ai/STT
I recommend asking there, they have a very active matrix channel and a forum on GitHub.