ElevenLabs launches Scribe v3 and Scribe v3 Live speech-to-text models
Noisy multi-speaker accuracy and sentence-boundary detection address the failure points that make transcripts from Whisper or basic APIs need heavy cleanup, reducing post-editing work. Developers building meeting notetakers, call-centre analytics or live captioning should evaluate Scribe v3 Live against AssemblyAI and Deepgram on their own audio. ElevenLabs is a US-based lab, so this widens vendor choice for teams globally without regional pricing or availability constraints announced.
ElevenLabs has released Scribe v3 and Scribe v3 Live, its newest speech-to-text models, which the company describes as its most accurate to date. The launch was announced via ElevenLabs' official account, with both models positioned for real-world audio conditions rather than clean studio recordings.
According to the announcement, Scribe v3 is designed to handle difficult listening environments, maintaining accuracy when audio contains extreme background noise. The model can distinguish pauses from the end of sentences, improving punctuation and segmentation, and performs speaker diarization, keeping track of who said what in a busy multi-person meeting. Scribe v3 Live is the streaming variant of the same model family, enabling real-time transcription use cases.
Speech-to-text has become a competitive layer of the AI stack, with rivals including OpenAI Whisper, AssemblyAI and Deepgram serving transcription-heavy workflows in meetings, call centres and media. Accuracy on noisy, multi-speaker audio is a key differentiator for enterprise buyers, and ElevenLabs' move extends its audio platform beyond voice generation into full speech recognition infrastructure.
Introducing Scribe v3 and Scribe v3 Live: our most accurate Speech to Text models yet. Built for real-world audio, Scribe excels in situations with extreme background noise, knows pauses from the end of sentences, and keeps track of who said what in a busy meeting.
This report is categorized under AITOP's "Products" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include ElevenLabs, Scribe v3, speech-to-text, multimodal.
Source attribution: curated from ElevenLabs (@elevenlabsio).
