Anuvani-LID: Robust and Calibrated Spoken Language Identification for Indian Languages

Key points

• Every multilingual speech pipeline begins with a language decision, an error there propagates silently through recognition, translation and routing.1
• Foundational speech models increasingly present language identification (LID) as an inherent latent, generalized capability.1
• Autoregressive decoders in speech models emit a special language ID token, and modern LLMs with audio capabilities can name the language when prompted.1
• We test this assumption on identical clips and find that whisper and similar speech models has tokens for only 15 of the 23 languages, on one second inputs Gemini 3.5 Flash reaches 0.52 macro-F1 and GPT-audio 0.14.1

Sources (1)

• [1]Anuvani-LID: Robust and Calibrated Spoken Language Identification for Indian Languages
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 14, 02:29 PM

  We present Anuvani LID, a dedicated 115M parameter LID conformer model trained on 54,643 hours of Indian speech spanning 22 Indian languages and Indian accented English.

  Every multilingual speech pipeline begins with a language decision, an error there propagates silently through recognition, translation and routing.

Extractive summary: sentences quoted from the sources.