AirCaps A5S Async v1 tops Voice Arena diarization+ASR board at 23.89% cpWER
rdesh26 · x · 2026-11-10
AirCaps A5S Async v1 takes the #1 spot on the Voice Arena Diarization + ASR leaderboard.
- Evaluated alongside 14 other systems on the same 139 held-out conversations (2–8 speakers, overlapping speech counted): 23.89% cpWER overall, with 24.91% on in-person chats and 17.47% on online calls — best in both settings.
- Runner-up Nemotron 3.5 Transcribe misses a slightly smaller share of words (8.32% vs 9.29%), but AirCaps is notably better at assigning words to the right speakers.
- A commenter suggests stress-testing it on CHiME-6 real-world multi-speaker data.
Related event: AirCaps A5S Async v1 Tops VoiceArena Speech Recognition Leaderboard(3 posts)→
More from Models
- Dev spends 2 hours replicating a Fable prompt with Opus; original took Fable 10 minutes — teortaxesTex · 2026-11-10
- NVIDIA's Jacob Loewenstein: token efficiency matters more than price sheets when picking models — togethercompute · 2026-11-10
- Vals AI: high effort beats max on Claude Fable 5.2, and max costs 5x more — airesearch12 · 2026-11-10
- Fable 5.2 builds a complete playable mobile game, Slingsong, from a single prompt — AIandDesign · 2026-11-10
- First impressions of Fable 5.2: constant session compaction, extreme slowness, task refusal — ersatzben · 2026-11-10
- Claude solved a 23-year-old black hole calculation and two open theoretical physics problems — claudeai · 2026-11-10