SEA-SpeechBench - Age Recognition (SEA Prompt): leaderboard
Metric: Macro-F1 (%; speaker age group (teens, adults, seniors) 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: Around 2028. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen3 Omni 30B A3B Instruct | 36.7 |
| 2 | Gemini 2.5 Flash | 36.33 |
| 3 | GPT-4o Audio | 34.27 |
| 4 | Gemma 3n E4B (IT) | 29.97 |
| 5 | Voxtral-Mini-3B-2507 | 28.49 |
| 6 | Phi-4 Multimodal Instruct | 28.42 |
| 7 | gemma-3n-E2B-it | 22.7 |
| 8 | Qwen2.5-Omni-7B | 18.28 |
| 9 | SeaLLMs-Audio-7B | 18.08 |
| 10 | Qwen2-Audio-7B-Instruct | 15.39 |
Interactive version: theaggregate.ai/benchmark?slug=sea-speechbench-age-recognition-sea-prompt · How It Works · Data refreshed daily, snapshot 2026-09-26.