CIIL's multilingual speech stack: open models for all 22 scheduled languages of India and English.
-
CIIL/Anuvani-LID
Audio Classification • 0.1B • Updated • 175 • 5 -
CIIL/Anuvani-Diarization
Voice Activity Detection • 76.1M • Updated • 42 • 1 -
Anuvani-LID: Robust and Calibrated Spoken Language Identification for Indian Languages
Paper • 2610.20263 • Published -
Anuvani-Diarization: Zero Shot Speaker Diarization and Annotation Conventions Across 22 Languages
Paper • 2610.20327 • Published