livingingroups/animal2vec
animal2vec: A self-supervised transformer for rare-event raw audio input
This project helps bioacoustic researchers analyze large audio datasets for rare animal vocalizations. It takes raw, unlabeled audio recordings, learns patterns from them, and then uses a small amount of labeled data to identify specific animal calls. Scientists, conservationists, and ecologists can use this to efficiently process vast amounts of acoustic data to understand animal behavior or monitor populations.
Use this if you need to detect infrequent animal sounds within extensive audio recordings, especially when you have limited labeled examples for training.
Not ideal if you are working with non-audio data, or if your primary goal is real-time detection on edge devices with very limited computational resources.
Stars
30
Forks
7
Language
Python
License
MIT
Category
Last pushed
Dec 15, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/livingingroups/animal2vec"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
woheller69/whoBIRD
Identify bird sounds in real time with this Android version of BirdNET. Bird sound recognition...
jhuus/HawkEars1
⚠️ HawkEars 1.0 (obsolete). See HawkEars 2.0 → https://github.com/jhuus/HawkEars
kahst/BirdNET-Electron
Electron app for sound file analysis with BirdNET.
meteyaran/Ethno-Ornithology-Atlas
Where avian impressions and linguistic traces converge. An interdisciplinary platform blending...
Komorebirumu/awe-ms-20260317-0957-02
AI Birdsong ID & Behavior Analyzer Code | MS-20260317-0957-02