https://github.com/nitely/speech-to-text

Speech-To-Text experiments
https://github.com/nitely/speech-to-text

Last synced: 28 days ago
JSON representation

Speech-To-Text experiments

• Host: GitHub
• URL: https://github.com/nitely/speech-to-text
• Owner: nitely
• Created: 2024-12-01T00:09:48.000Z (about 1 month ago)
• Default Branch: master
• Last Pushed: 2024-12-01T06:47:34.000Z (about 1 month ago)
• Last Synced: 2024-12-01T07:29:03.703Z (about 1 month ago)
• Language: Nim
• Size: 63.5 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
Speech-To-Text experiments; nothing to see here; shhhh

## What?

- Get audio from device (mic or desktop) or audio file

- Normalize the audio to 16kHz mono

- Get the speech segments with VAD with a min/max duration

- Get the transcript of the segment with Whisper

- Ideally with real-time stream support. Albeit whisper.cpp works

  best on 30s segments

## What for?

- Voice comms

- Transcriptions

- Closed Caption

- Real-time translation

- Capture system output speech to text

## Why?

Learnings.

## License

Unlicense. You cannot use this or even look at it. Move along.