Audio2Tool - Correction: leaderboard

Speech-driven tool calling on Audio2Tool (2026), smart car, smart home and wearable requests spoken by diverse synthetic voices with background noise and given to an end-to-end speech language model, on the correction tier (requests revised mid-utterance). Exact match of the tool calls and all arguments, in percent; higher is better.

Metric: Exact match (%): share of queries whose predicted tool or ordered tool sequence and all arguments exactly match the reference after deterministic normalization, on the 4,560 Tier 6 correction queries (mid-utterance revisions of the request) of Audio2Tool, with the spoken query (voice-cloned TTS speech mixed with background noise) given to an end-to-end speech language model, zero-shot; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 5 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen2.5-Omni-7B42#1888
2Kimi-Audio39.1#1722
3Step-Audio-2-mini33.4
4Qwen 3 Omni 30B A3B24.8#851
5Audio Flamingo 321.6

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=audio2tool-correction · How It Works · Data refreshed daily, snapshot 2026-11-22.