xinjli/allosaurus

Allosaurus is a pretrained universal phone recognizer for more than 2000 languages

65
/ 100
Established

Built on a multilingual allophone system architecture from ICASSP 2020 research, it uses pretrained acoustic models that can be swapped (universal or language-specific) and supports language-specific phone inventory constraints to improve accuracy. Provides both command-line and Python APIs with automatic audio resampling, GPU acceleration support, and downloadable model variants—including a universal model covering 2000+ languages and specialized models for individual languages like English.

715 stars and 7,882 monthly downloads. No commits in the last 6 months. Available on PyPI.

Stale 6m
Maintenance 0 / 25
Adoption 19 / 25
Maturity 25 / 25
Community 21 / 25

How are scores calculated?

Stars

715

Forks

100

Language

Python

License

GPL-3.0

Last pushed

Apr 26, 2024

Monthly downloads

7,882

Commits (30d)

0

Dependencies

6

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/xinjli/allosaurus"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.