Salesforce's SOV-Transcribe-1 hits 60x speed factor in Artificial Analysis ASR leaderboard
ArtificialAnlys · x · 2026-11-13
Artificial Analysis updates its Speech-to-Text leaderboard: SOV-Transcribe-1 on Salesforce's hosted endpoint reaches a 60x Speed Factor, transcribing 1 minute of audio in about 1 second.
Key points:
- Accuracy measured by AA-WER v2, an audio-duration-weighted average over 8 hours from AA-AgentTalk (50%), VoxPopuli (25%), and Earnings22 (25%);
- Also compares Speed Factor and price per 1000 minutes;
- 81 models ranked, open weights and proprietary; time-limited models are chunked to 9 min or 30 s;
- Full results, methodology, and model weights on Hugging Face are public.
Related event: Salesforce Open-Sources SOV-Transcribe-1 with 60x Transcription Speed(2 posts)→
More from Models
- Ditto rolls out uncensored inference to all users — bittingthembits · 2026-11-13
- Google says Gemini 4 Hydrogen expands query fan-out and the range of sites Search links to — OnlineInference · 2026-11-13
- 124B model in 22GB: Cactus Quants hits 1.418 bits/weight, 47 tok/s on an M5 Max — pbaylies · 2026-11-13
- PostTrainBench: Fable 5.2 ties 5.1 at 44.6%, speculating Anthropic capped AI research ability — scaling01 · 2026-11-13
- Gemini 4 Hydrogen lands in Devin, scoring 38.2% on FrontierCode 2.0 behind GPT-6.1 Sol — scaling01 · 2026-11-13
- Gemini 4 Carbon comes within 1.9 points of Fable 5.2 on Code4Scene at ~1/10 the cost — ZhitingHu · 2026-11-13