DeepSpeech 60x Smaller, 9x faster, and 2x accuracy

Re:DeepSpeech 60x Smaller, 9x faster, and 2x accuracy by colechristensen on Wednesday March 09, 2022 @12:18PM (#30617182)

What are the state of the art open solutions to local voice recognition? Preferably with available models that a small org can also train themselves without millions in hardware.

[ Parent | on HN ]

Re:DeepSpeech 60x Smaller, 9x faster, and 2x accuracy by rileyphone on Wednesday March 09, 2022 @01:37PM (#30618316)

I will add https://github.com/coqui-ai/STT, which is a continuation of DeepSpeech. Also, I've been messing around with https://github.com/ideasman42/nerd-dictation, which works on a VOSK backend - accuracy is decent, especially with the bigger model.

[ Parent | on HN ]

Re:DeepSpeech 60x Smaller, 9x faster, and 2x accuracy by trowngon on Wednesday March 09, 2022 @01:48PM (#30618465)

Vosk https://github.com/alphacep/vosk-api

[ Parent | on HN ]

Re:DeepSpeech 60x Smaller, 9x faster, and 2x accuracy by albertzeyer on Wednesday March 09, 2022 @12:29PM (#30617369)

Kaldi, K2, ESPNet.

[ Parent | on HN ]

Re:DeepSpeech 60x Smaller, 9x faster, and 2x accuracy by misc1234 on Wednesday March 09, 2022 @12:50PM (#30617692)

Nemo

[ Parent | on HN ]