Audio Chunker Tool

A Python CLI tool to split audio files into chunks based on silence detection. Perfect for preparing datasets for machine learning (TTS, ASR) or simply breaking down long recordings.

Features

• Smart Splitting: Uses pydub's split_on_silence to find natural breaks in audio.
• Configurable: Adjust silence threshold, minimum silence length, and padding.
• Robust: Handles various audio formats (wav, mp3, flac, etc.) supported by ffmpeg.
• User-Friendly: Interactive CLI with progress bars and colorful output.

Prerequisites

1. Python 3.6+
2. FFmpeg: This tool requires FFmpeg for audio processing.
   • macOS: brew install ffmpeg
   • Ubuntu/Debian: sudo apt install ffmpeg
   • Windows: Download and add to PATH.

Installation

1. Clone this repository:

   git clone <your-repo-url>
   cd audio_chunker

2. Install Python dependencies:

   pip install -r requirements.txt

Usage

Usage

First, activate your virtual environment (if you haven't already):

source venv/bin/activate

Then run the script:

python chunker.py /path/to/your/audio_file.wav

Options

| Option | Flag | Default | Description | | hum | --- | --- | --- | | Output Directory | -o, --output-dir | chunks | Folder to save the output files. | | Min Silence Len | -m, --min-silence-len | 1000 | Minimum length of silence (ms) to split on. | | Silence Threshold | -t, --silence-thresh | -40 | Threshold (dBFS) to consider as silence. | | Keep Silence | -k, --keep-silence | 200 | Silence (ms) to keep at start/end of chunks. | | Prefix | -p, --prefix | chunk | Filename prefix for chunks (e.g., chunk_001.wav). | | Format | -f, --format | wav | Output format (wav, mp3, etc.). |

Examples

Basic split:

python chunker.py interview.mp3

Custom strictness (shorter silence allowed, quieter threshold):

python chunker.py podcast.wav -m 500 -t -50 -o output_folder