Voice of India - Assamese: leaderboard

Speech recognition on the Assamese part of Voice of India (2026), a closed benchmark of unscripted telephone conversations in Indian languages, scored with an orthographically informed word error rate that accepts permissible spelling variants. Percent, can exceed 100; lower is better.

Metric: Orthographically informed word error rate (OIWER, %; can exceed 100) on the Assamese unscripted telephone-conversation utterances of the closed Voice of India benchmark, transcripts with lattices of permissible spelling variants; systems run at default inference settings and only where they support the language; lower is better. Source: arxiv.org. Saturation forecast: Around March 2028. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Sarvam Audio12.7
2IndicConformer14.3
3elevenlabs-scribe_v215.6
4Gemini 3 Pro20.1#145
5omniASR_LLM_7B25.3
6Gemini 3 Flash26.9#175
7omniASR_LLM_1B29.2
8gpt-4o-mini-transcribe37.6
9GPT-4o Transcribe94.7
10AssemblyAI Universal104.8

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=voice-of-india-assamese · How It Works · Data refreshed daily, snapshot 2026-11-21.