
Scribe v3 by ElevenLabs精选
Speech to text with 27% fewer errors in 90+ languages
Scribe v3 is ElevenLabs’ most accurate speech-to-text model, for developers building on calls, meetings, interviews and media. It makes 27% fewer word errors than Scribe v2 across 90+ languages. Unlike transcription that breaks on noise, accents and crosstalk, it labels up to 32 speakers, tags laughter, music and emotion, boosts up to 1,000 of your own terms, detects 56 entity types and handles files up to 15 hours. For live use, Scribe v3 Live runs under 150ms.
2026-11-11 同榜
801
ScroogeFrog AIAI ad fraud detection that protects your marketing budget ▲ 30002
Nod by n8nOpen source personal agent with its own computer▲ 29203
CALL-EMake phone calls with your AI agent (Muse, ChatGPT, Claude)▲ 25105
SkillboxOne library for your AI skills, synced to every agent▲ 15206
MiniMax H3-ImageBringing 4K image generation into the 10-second era▲ 12707
RelayDocSend for AI agents: send and receive docs with context▲ 12608
Shot.videoMulti-camera videos with the phones you already own▲ 11709
Reddit ReviewsProduct pages that collect subreddit sentiment▲ 112