Muse Voice Transcribe

Competitive latency and streaming transcription accuracy with live attribution for 20+ speakers. Build with a single model at $0.18/hour.

Start building

Documentation

Meet Muse Voice Transcribe

Competitive transcription accuracy

Leading streaming word error rate, and it holds that accuracy on your vocabulary through contextual and keyword biasing, with no fine-tuning.

Live speaker and turn awareness

Speaker attribution for 20+ speakers and end-of-speech detection happen inside the recognition model, not in a batch pass at the end of the audio stream.

Production pricing

Ship with leading transcription accuracy and streaming capabilities without compromising on costs.

Muse Voice Transcribe benchmarks

Streaming STT WER leaderboard - image2.png

AA-WER Streaming Index - Final Transcription

Muse Voice Transcribe pricing

Model

Price / 1,000 min

Price / hour

muse-voice-transcribe-1.0

$3.00

$0.18

Build with Muse Voice Transcribe

Start building

Cookbooks and quickstarts

Read documentation

Quick startYour first working request in under five minutes. Point your existing OpenAI SDK compatible client at Meta Model API.

Learn more

Voice API FundamentalsGet started and learn how to handle punctuation, speech-boundary detection and speaker attribution.

Learn more

Stay up-to-date

Our latest updates delivered to your inbox

Subscribe to our newsletter to keep up with the latest AI updates, releases and more.