Edge0 / ARK-ASR-0.6B

Audio1.3B44058Updated Jun 27, 2026

Downloads

440

Likes

58

Last updated

Jun 27, 2026

License

apache-2.0

TL;DR ARK-ASR-0.6B is an automatic speech recognition model trained with teacher-data adaptation and on-policy distillation, using a compact 0.6B-scale decoder LLM together with a dedicated audio encoder and adapter. The accompanying training, inference, and evaluation code i…

Tasks & library

  • automatic speech recognition
  • transformers

Tags

  • transformers
  • arkasr
  • text-generation
  • speech
  • audio
  • pytorch

Technical specifications

Parameters1.3B
Librarytransformers
Languages19 languages
Licenseapache-2.0