View Code? Open in Web Editor
NEW
License: GNU General Public License v3.0
searchoney's Introduction
- All transcriptions are copyright OneyPlays, O'Neil et. al, 2014-23.
- Techstack highly inspired by searchTAFS.
- Thanks to Marceline for conceptual design and theory.
- Compressed YouTube audio size:
71GB
- Uncompressed 16KHz audio size:
329GB
- Scraping & conversion: yt-dlp, ffmpeg
- Speech processing: MacWhisper and its
base.en.bin
model, done on my M2 8/256 MacBook Air
- Transcripts are currently in semicolon-formatted CSV, but will be converted to normal JSON upon completion.
searchoney's People
Contributors
Watchers