Maybloc: SOTA speech-to-text for multi-speaker audio
New transcription API (model: Calico) launched Oct 15 on Product Hunt, claiming the lowest diarization error on NOTSOFAR/AMI/DiPCo — 45 languages, up to 25 speakers, overlapping speech kept, $0.006/min, add-ons: subtitles, PII redaction, speaker names.
Why it matters
Purpose-built for messy multi-voice audio where single-speaker tools drop overlapping speech; claims 40%+ fewer errors than AssemblyAI and pyannote on real office meetings.
What you could build with it
A sales-coaching startup could run Maybloc over multi-speaker call recordings to attribute every sentence and catch overlaps, producing per-rep talk-time analytics that single-speaker transcription tools miss.
Does it hold up?
Too early to judge — launched Oct 15 with vendor-reported diarization benchmarks; no independent user reports yet.
Built with Maybloc
- Maybloc: new SOTA for speech-to-text for multi-speaker audio — Product Hunt launchproducthunt · The team's own launch post (Oct 15, 2026): Calico model claims lowest diarization error vs pyannote/AssemblyAI/Deepgram on NOTSOFAR, AMI, DiPCo; 45 languages, 25 speakers, 3-hour files, $0.006/min; invites users to test on their messiest recordings.
- Maybloc Reviews on Product Huntproducthunt · Early user reviews and ratings from the launch week — the first real user feedback on the product beyond the team's own claims.
- Maybloc benchmark results — official benchmarks pagemaybloc · Official benchmark tables: 43% fewer errors than AssemblyAI/pyannote on NOTSOFAR office meetings, 54–73% fewer than Deepgram Nova-3, best published result on AMI ahead of Gemini and VibeVoice; plus pricing, API, and add-on details.
Learn more
First spotted on article: source.
More AI models
OpenAI publishes 722 AI-generated math manuscripts from an unreleased modelOpenAI released 722 mathematical manuscripts in 372 result families to GitHub under Apache-2.0, produced by an…models · JEV 0.74Tencent Hunyuan Hy3: open 295B MoE with 21B active under Apache 2.0Tencent open-sourced Hy3, a 295B-parameter mixture-of-experts model with only 21B parameters active per token, under…models · JEV 0.73NVIDIA Isaac Lab 3.0 GA + Isaac OS open runtime + Agile ONE SN humanoidMajor Isaac platform expansion at GTC Berlin Oct 21: Isaac Lab 3.0 is generally available as the open-source robot…models · JEV 0.73StepFun Step-5-Preview — open BF16 weights releasedStepFun published the full BF16 weights for its flagship Step-5-Preview agentic model (600B-parameter sparse MoE, 27B…models · JEV 0.73NVIDIA Nemotron 3.5 Retriever collection: Embed 8B/1B, Rerank 8B, Parse 631MNVIDIA's Oct 21 collection drop for agentic retrieval: Nemotron-3.5-Embed in 8B BF16, 1B BF16 and 1B NVFP4…models · JEV 0.72Kimi K3.1 Pro: Moonshot's 2.8T open-weight flagship officially launchesMoonshot AI officially launched Kimi K3.1 Pro on Oct 23, a mixture-of-experts model with 2.8 trillion total / 104…models · JEV 0.72
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.