Benchmark Almanac

Indic DiarBench

Sarvam 16% tops its leaderboard GPT-4o Mar 2025 36.2% leads the models we track
Not in model cards

Models transcribe conversations in 22 Indian languages and identify who spoke each word. Human annotators supply speaker labels and timestamps, including for overlapping speech.

Sarvam and AI4Bharat report three error rates for eight systems. The test draws 1,164 clips from 590 recordings. Language coverage varies by system.

Released
July 2026
Kind
Speech recognition and translation
Level
Everyday
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
Many languages
In and out
Audio → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Gemini 3 Pro #56 on the capability index
Last model added
November 2025
Scores refreshed
November 20th, 2026
8
Entries in its paper, 2 models we track

Sarvam tops the leaderboard with 16% GPT-4o leads with 36.2%

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is GPT-4o, at 36.2%. This view shows only models we track, each at its best.

  1. 1 Sarvam 16%
  2. 2 AWS Transcribe 23.5%
  3. 3 Deepgram Nova-3 32%
  4. 4 Azure STT 34.8%
  5. 5 ElevenLabs Scribe 35%
  6. 6 GPT-4o Mar 2025 36.2%
  7. 7 AssemblyAI 40.5%
  8. 8 Gemini 3 Pro 74%

Source: The Indic DiarBench paper

Source: The Indic DiarBench paper

#EntryScoreSettingDateSource
1 Sarvam 16% Official leaderboard
2 AWS Transcribe 23.5% Official leaderboard
3 Deepgram Nova-3 32% Official leaderboard
4 Azure STT 34.8% Official leaderboard
5 ElevenLabs Scribe 35% Official leaderboard
6 GPT-4o Mar 2025 36.2% Mar 2025 Official leaderboard
7 AssemblyAI 40.5% Official leaderboard
8 Gemini 3 Pro 74% Nov 2025 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-4o Mar 2025 36.2% Mar 2025 Official leaderboard
2 Gemini 3 Pro 74% Nov 2025 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

In the benchmark’s paper
0% 20% 40% 60% 80% 100% 2025 2026 GPT-4o 0% 20% 40% 60% 80% 100% 2025 2026 GPT-4o

Source: Official Indic DiarBench leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Audio and video

All 695 in Audio and video
  • Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy. Qwen 3.8 Max 88.7%
  • LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization. Gemini 4 Argon 91.7%
  • MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy. Qwen 3.8 Max 82.4%
  • MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy. Qwen 3.8 Max 90.8%
  • FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better. Phi-4-multimodal 4.00%
  • VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules. Qwen3.5-Omni-Plus 93.1%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 12 months ago, in November 2025, for Gemini 3 Pro. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the benchmark’s paper (8 entries).
  • Pass Models scored 2 models we track, plus 6 other entries on its leaderboard.
  • Warning Newest score The newest came 12 months ago, in November 2025, for Gemini 3 Pro.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3 Pro (#56).
  • Note Headroom Lower is better, so there’s no ceiling to reach.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 2 rows for models we track and 6 other entries at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check