They clip what you said. We clip what you felt.
MoodCutter is an open-source proof of concept for an AI system that automatically detects and extracts emotionally resonant moments from video — funny, sad, and high-energy — to help content creators find their best clips in seconds, not hours.
Content creators spend hours scrubbing through footage to find the 3 seconds that made someone laugh or cry. Existing AI clippers analyze transcripts and content patterns — they detect what you said, not how it made people feel. MoodCutter is being built to close that gap.
The proof of concept has achieved its goal of validating audio-based emotional moment detection. Active development has moved to the full product, which is being built privately.
This repository is archived as a reference implementation. No further features will be added here.
The v0.1 implementation uses audio-based loudness detection as a foundation for emotional moment detection in video.
- ✅ Audio loudness analysis using RMS energy
- ✅ Dynamic threshold-based loud moment detection
- ✅ Time-stamped moment extraction
- ✅ Energy visualization graph
- ✅ Waveform analysis with highlighted segments
- ✅ Command-line interface
Audio analysis results showing detected loud moments with timestamps
Bar chart showing audio energy levels over time with detection threshold
Audio waveform with highlighted loud segments
git clone https://github.com/axtheon/MoodCutter.git
cd MoodCutter
pip install -r requirements.txtpython src/CLI/main.pyThe script will:
- Load the test audio file (
examples/audio/test_audio.wav) - Analyze audio in 0.5-second chunks
- Detect moments exceeding 1.5× average energy
- Display timestamps in terminal
- Show an interactive energy visualization graph
CHUNK_DURATION = 0.5 # Analysis window size (seconds)
THRESHOLD_MULTIPLIER = 1.5 # Sensitivity (lower = more sensitive)- Audio Loading — Loads audio using
librosa - Chunking — Divides audio into fixed-duration segments
- Energy Calculation — Computes RMS energy (loudness) per chunk
- Threshold Detection — Flags chunks exceeding the dynamic threshold
- Visualization — Displays energy levels and detected moments
- Python
- librosa — audio analysis
- matplotlib — visualization
This repository covers the foundational proof of concept only — audio energy detection as a single signal. Further development is closed-source.
The full product is being built as a multimodal emotion detection system for content creators.
This repository is published as a reference proof of concept. It demonstrates that automated emotional moment detection in audio is technically feasible.
The full product is in private development. This POC is not actively maintained.
MIT License — see LICENSE for details.
Built by @axtheon — follow for updates on the full product.