Voxtral Mini Transcribe 3
Voxtral Mini Transcribe 3 is a speech-to-text model from Mistral AI that converts audio recordings into written transcripts through an API. Developers can submit audio for batch transcription and receive text with timestamps and speaker labels. It supports multilingual transcription, speaker diarization, and context biasing, which lets an application supply names or specialized vocabulary to guide recognition. The model is designed for developers building applications and audio workflows that need to make recorded speech searchable, attribute words to speakers, or align transcript segments with audio. Teams can call it through Mistral's hosted API and incorporate its output into their own software, such as transcription pipelines and voice-enabled services. It is a model/API offering rather than a standalone recording or editing application.
Learn more
Voxtral Small Transcribe 3
Voxtral Small Transcribe 3 is a multilingual speech-to-text model for transcribing recorded audio. Mistral provides it through an audio transcription API and documents timestamps, speaker diarization, and context biasing, which allows developers to supply key phrases to help recognize domain-specific terms and names. The model automatically detects supported languages or accepts a language code, and supports more than 50 languages. It is designed for application developers building recorded-audio workflows, including meeting, interview, and call transcription, customer-service systems, and voice-agent pipelines. Mistral also describes self-deployment options for the Voxtral 3 family, including private-cloud and on-premises hosting.
Learn more
Scribe
ElevenLabs has introduced Scribe, an advanced Automatic Speech Recognition (ASR) model designed to deliver highly accurate transcriptions across 99 languages. Scribe is engineered to handle diverse real-world audio scenarios, providing features such as word-level timestamps, speaker diarization, and audio-event tagging. Benchmark tests, including FLEURS and Common Voice, demonstrate Scribe's superior performance over leading models like Gemini 2.0 Flash, Whisper Large V3, and Deepgram Nova-3, achieving the lowest word error rates in languages such as Italian (98.7%) and English (96.7%). Notably, Scribe also significantly reduces errors in languages that have been traditionally underserved, including Serbian, Cantonese, and Malayalam, where other models often exhibit error rates exceeding 40%. Developers can integrate Scribe through ElevenLabs' speech-to-text API, receiving structured JSON transcripts that include detailed annotations.
Learn more
Voxtral Mini Transcribe Realtime 3
Voxtral Mini Transcribe Realtime 3 is a speech-to-text model delivered through Mistral’s API for applications that need live transcription. It streams audio over WebSocket and returns transcript events while a person is speaking, with configurable streaming delays. Developers can use it to add real-time captions, voice-assistant input, or transcription to voice-agent workflows. The model provides turn detection, speaker diarization, and context biasing for domain-specific vocabulary, and supports transcription across 45 languages and dialects. Applications can connect to its real-time endpoint and receive transcription events as audio arrives, rather than waiting for an entire recording to finish. Voxtral Mini Transcribe Realtime 3 is part of Mistral’s Voxtral audio model family.
Learn more
