# Omi Medical Speech-to-Text documentation

> Public developer documentation for the Omi Medical STT API.

## Start here

- [Overview](https://console.omi.health/docs): Choose synchronous, asynchronous, or realtime transcription and run a first request.
- [Synchronous transcription](https://console.omi.health/docs/transcription): Multipart fields, response formats, models, and examples.
- [Asynchronous transcription](https://console.omi.health/docs/async): Direct-upload 202 dispatch, presigned large uploads, polling, results, idempotency, and retention.
- [Realtime preview](https://console.omi.health/docs/realtime): WebSocket start/audio/stop flow and draft-versus-record authority.

## Features

- [Languages](https://console.omi.health/docs/languages): en/es/pt/fr/de/nl/ar/hi, regional aliases, explicit language selection, and per-utterance language=auto.
- [Vocabulary and Patterns](https://console.omi.health/docs/context): Encounter term lists, stored dictionaries, safety fallback, and request-scoped structured-token patterns.
- [Speakers and timestamps](https://console.omi.health/docs/speakers): Speaker diarization and acoustic word timestamps without transcript mutation.
- [Webhooks](https://console.omi.health/docs/webhooks): Standard Webhooks signatures, retries, destination safety, and secret rotation.

## Reference

- [API reference](https://console.omi.health/docs/reference): Authentication, endpoints, every public field, schemas, limits, errors, retries, and data handling.
- [OpenAPI 3.1](https://console.omi.health/openapi.yaml): Curated customer-facing HTTP contract.
- [Complete Markdown guide](https://console.omi.health/docs.md): Compact machine-readable documentation.
- [Service status](https://d14wqvek28lca6.cloudfront.net/): Independent five-minute probes for batch and realtime.

## Important contract notes

- Base URL: https://api.omi.health
- Authorization: Bearer <OMI_API_KEY>
- Omi Medical STT is an English-first medical speech model.
- English is the benchmarked and medically evaluated language.
- es/pt/fr/de/nl/ar/hi were included in training and are available for testing and evaluation; they are not medically evaluated and should not be relied on for medical accuracy.
- Automatic language detection is opt-in with language=auto; omitted language does not enable detection.
- Batch transcription is generally available to all API keys.
- The direct-upload front door accepts up to exactly 100,000,000 bytes and returns either inline 200 or async 202.
- Files of 30 seconds or longer use the asynchronous-optimized pipeline.
- Files above the direct cap use the presigned large-upload path up to 1 GiB.
- Realtime is an invited preview and is not open to all accounts. Async language=auto and Patterns are also invited preview features.
- Realtime drafts are non-authoritative. Only the tier=record, is_final=true event is the official record.
- Published benchmark claims apply to English.