SEA-SpeechBench - Gender Recognition (SEA Prompt): leaderboard
Speaker gender recognition in SEA-SpeechBench (Liao et al., 2026) from short clips in 11 Southeast Asian languages, with the instruction written in the native Southeast Asian language. Macro-F1 in percent after mapping answers to labels; higher is better.
Metric: Macro-F1 (%; speaker gender from speech, labels canonicalised by bilingual string matching, per-dataset scores averaged; short clips of at most 30 s from curated public Southeast Asian speech corpora, up to 1,000 sampled items per dataset; instruction prompt in the native Southeast Asian language). Source: arxiv.org. Saturation forecast: Estimated already saturated. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash | 82.54 | #452 |
| 2 | Qwen3 Omni 30B A3B Instruct | 65.43 | #902 |
| 3 | Qwen2.5-Omni-7B | 49.17 | #1797 |
| 4 | MERaLiON-2-10B | 45.93 | #1646 |
| 5 | MERaLiON-2-3B | 36.48 | #6258 |
| 6 | Phi-4 Multimodal Instruct | 31.06 | #3914 |
| 7 | Gemma 3n E4B (IT) | 23.54 | #2220 |
| 8 | Voxtral-Mini-3B-2507 | 20.41 | #3702 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=sea-speechbench-gender-recognition-sea-prompt · How It Works · Data refreshed daily, snapshot 2026-11-12.