Scribe v2
ElevenLabs' state-of-the-art speech recognition model, delivering highly accurate transcription across 90+ languages with advanced features like speaker diarization, entity detection, and precise word-level timestamps.
Multilingual speech transcription with speaker diarization
Scribe v2 is ElevenLabs' flagship speech-to-text model, built to transcribe audio accurately across more than 90 languages with automatic language detection. It supports speaker diarization for up to 32 speakers, word-level timestamps, and entity detection across 56 named entity types, making it one of the more feature-rich transcription models available through an API. Developers can also supply up to 100 custom keyterms to improve recognition of domain-specific vocabulary, names, or technical jargon.
Scribe v2 is well suited for applications where transcription accuracy and rich metadata matter — such as meeting summarization, podcast indexing, media subtitling, and legal or medical documentation workflows. Its dynamic audio tagging feature automatically labels non-speech events, which adds context beyond spoken words. The combination of precise timing data and speaker attribution makes it a practical choice for any pipeline where knowing who said what and when is a requirement.
What Scribe v2 supports
Multilingual Transcription
Transcribes spoken audio in over 90 languages with automatic language detection, requiring no manual language configuration.
Speaker Diarization
Identifies and separates individual speakers within a single audio file, supporting up to 32 distinct speakers.
Word-Level Timestamps
Provides precise timing for every transcribed word, enabling accurate alignment with audio or video content.
Entity Detection
Automatically identifies and labels named entities within transcriptions, covering up to 56 entity types.
Keyterm Prompting
Accepts up to 100 custom keyterms to guide the model toward accurate recognition of domain-specific vocabulary or proper nouns.
Audio Event Tagging
Detects and labels non-speech audio events dynamically, adding contextual metadata beyond spoken words.
Ready to build with Scribe v2?
Get Started FreeCommon questions about Scribe v2
How many languages does Scribe v2 support?
Scribe v2 supports transcription in over 90 languages and can automatically detect the spoken language without requiring manual configuration.
Does Scribe v2 have a context window limit?
No context window is specified in the available metadata for Scribe v2, as it is a speech-to-text model rather than a text-based language model. Limits, if any, would apply to audio file length or size as defined by the ElevenLabs API.
How many speakers can Scribe v2 distinguish in a single file?
Scribe v2's speaker diarization feature can identify and separate up to 32 individual speakers within a single audio file.
Can I improve recognition of specialized terminology?
Yes. Scribe v2 supports keyterm prompting, which allows you to supply up to 100 custom terms — such as product names, technical jargon, or proper nouns — to guide the model toward more accurate recognition.
What types of named entities can Scribe v2 detect?
Scribe v2 can automatically identify and label up to 56 types of named entities within a transcription, such as people, organizations, and locations.
Parameters & options
Choose whether to include timing and speaker information in the transcription
Explore similar models
ElevenLabs Scribe v1
ElevenLabs' first-generation speech-to-text transcription model, offering accurate audio transcription across multiple languages.
ElevenLabs ElevenLabs TTS
ElevenLabs is a leading voice AI platform that generates emotionally rich, human-like speech across 70+ languages with advanced voice cloning and real-time conversational AI capabilities.
ElevenLabs ElevenLabs Music
music_v1 is an AI model designed for music-related generation or analysis tasks.
OpenAI GPT-4o mini Transcribe
Speech-to-text model powered by GPT-4o mini
Google Gemini 3.1 Pro
Google's frontier reasoning model delivering enhanced software engineering, agentic reliability, and multimodal intelligence across a 1M-token context window.
Black Forest Labs FLUX.2 [max]
Black Forest Labs' most powerful image generation and editing model, delivering professional-grade output with real-time web context and unmatched visual consistency.
Start building with Scribe v2
No API keys required. Create AI-powered workflows with Scribe v2 in minutes — free.