What are the latest Open Source Speech To Text Models with a focus on real-time

Models
1

Hey, do you know current models that can also be executed locally, i.e. not in the cloud

2

When it comes to locally executable models, the Whisper series seems to have a lot of know-how. However, there are other options as well.

In terms of speed, FastRTC excels in real-time performance, but it’s quite specialized. Or rather, it’s cloud-based?

3

Yes, I already have Whisper on my shortlist and it seems to be the best option. I’ve also heard about

  • Kaldi
  • DeepSpeech
  • Vosk
  • SpeechBrain

Do you have any experience with these?

4

Do you have any experience with these?

No.