In essence, this script creates a full-featured local desktop utility to:
- Transcribe an audio file into individual words.
- Display and interact with each word’s start and end positions on a timeline or within the "Review Dashboard."
- Adjust timing offsets for the beginning and end of each word either globally or individually.
- Play full audio or specific words directly from within the app.
- Export words as separate `.wav` audio files.
GitHub repository: https://github.com/Northstrix/bootleg-text-slicer
Successfully tested with English and Italian audio files.
AMD Athlon 3050U Performance:
Track: alicenelpaesemeraviglie_04_carroll_64kb.mp3 (https://librivox.org/le-avventure-dalice-nel-paese-delle-meraviglie-by-lewis-carroll/)
Track Duration: 1027.60 seconds (~17:08 min)
Words Detected: 2469
Processing Time: 1812.69 seconds (~30:13 min)
Efficiency: 0.57x Realtime
Made using Google AI Studio (Gemini 3 Flash Preview)
Bootleg Text Slicer
Text transcription & slicing tool with visual timeline and WAV output.
Brought to you by:
northstrix
Downloads:
4 This Week