← Projects
AI Systems

Audio Splitter

A desktop GUI for visually splitting audio files with automatic Whisper transcription

Audio Splitter screenshot

Overview

Audio Splitter is a PySide6 desktop application for splitting audio files at user-defined points. Load an MP3 or WAV file and the full waveform renders immediately. Drag split markers freely to define exact cut positions — no snapping, full manual control.

Transcription runs automatically in the background using a local faster-whisper model the moment a file is loaded. No API key, no internet required after the initial model download. The transcript is saved as a .txt file alongside the audio.

Splitting uses ffmpeg stream copy — no re-encoding — so splits are instant and bit-perfect. Output always matches the input format. The linked chunks/duration controls let you set how many pieces you want or how long each should be, and markers redistribute automatically.

Tech Stack

PythonPySide6ffmpegWhisper
Not publicly deployed View Repo Support this project