AirCaps A5S Async v1 Benchmark - Agentic private meetings - cpWER: leaderboard

AirCaps A5S Async v1 Benchmark - Agentic private meetings - cpWER — AirCaps A5S Async v1 author comparison, November2026. Same 16kHz mono16bit WAV full recordings and native API defaults, with only exact model, language and diarization set. English language is en; Mandarin datasets use their native language. Meta switches from file to realtime above10min with same model/mode. Other documented request limits use fewest equal chunks, two-minute overlap and reference-blind stitching. Failed recordings are empty transcripts/all deletions. References and hypotheses receive OpenASR normalization plus ellipsis removal and split-numeral joining, per speaker runs, then MeetEval minimum-permutation error. Nvidia is BF16 self-hosted1xH100, temperature0,top_p1,seed0,reasoningoff and its published benchmark prompt; NVIDIA generation-loop guardrail retained. Model runs and native request settings are attached. AirCaps accuracy runNovember6; peersNovember1–2. This campaign is distinct from VoiceArena or another evaluator using the same corpus. Four licensed private online meetings from Agentic Data Company; machine-made references, data on request, scores only. Metric: cpWER (%). Source: https://github.com/aircaps/a5s-async-v1-benchmark

Metric: cpWER (%). Source: github.com. Saturation forecast: Rough model projection: around 2026. 9 models tracked. Updated 16 Nov 2026.

Top models

#ModelScoreOverall rank
1AirCaps A5S Async v18.46
2MAI-Transcribe-210.1
3Nemotron 3.5 Transcribe13.35
4Scribe v214.96
5Muse Voice Transcribe 1.015.4
6Gemini 3.5 Transcribe17.14
7AssemblyAI Universal-3.5 Pro18.98
8Deepgram Nova 325.54
9Grok Voice Transcribe 2.034.08

Interactive version: theaggregate.ai/benchmark?slug=aircaps-a5s-async-v1-benchmark-agentic-private-meetings-cpwer · How It Works · Data refreshed daily, snapshot 2026-11-16.