Speech-to-Text
Speech-to-Text
Transcribe live and recorded speech in 60+ languages, with speaker labels, language identification, and real-time translation.
Soniox Speech-to-Text turns speech into text in 60+ languages. Stream live audio over a WebSocket and get words back as people speak, or send recorded files and get a full transcript. Languages can switch mid-sentence, and Soniox keeps up.
Get started
Run your first transcription in a few minutes.
Real-time API
Stream live audio and get text back while people speak.
Async API
Transcribe recorded files from a URL or an upload.
Speaker diarization
Know who said what in multi-speaker audio.
Language identification
Detect the spoken language of every part of the audio.
Models
Pick the model for your latency and accuracy needs.