IS2AI/TurkicASR

A multilingual ASR model that can recognize ten Turkic languages—Azerbaijani, Bashkir, Chuvash, Kazakh, Kyrgyz, Sakha, Tatar, Turkish, Uyghur, and Uzbek.

41
/ 100
Emerging

Built on ESPnet, the project combines multiple public datasets (KSC, TSC, USC, and Common Voice 10.0) to train multilingual end-to-end ASR models, with options for Turkic-only or polyglot training. Pre-trained models are provided for immediate inference on 16kHz WAV files via a simple command-line interface. The architecture supports both single-language and cross-lingual training strategies, enabling transfer learning across related Turkic language families.

No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

82

Forks

11

Language

Python

License

CC-BY-4.0

Last pushed

Aug 01, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/IS2AI/TurkicASR"

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