Suisse Speech API documentation

Base URL https://api.suisse-speech.ch/v1 · X-API-Key

Speech-to-text and text-to-speech for Swiss German, its regional dialects and other languages. Both directions are available in two modes: batch for whole files and whole texts, realtime for streaming over a WebSocket.

Base URL: https://api.suisse-speech.ch/v1
Authentication: an X-API-Key header on every request (Authentication)
Reference: API reference, generated from the OpenAPI 3.0 description (openapi.json)
Machine-readable capabilities: GET /v1/capabilities

This documentation is the narrative guide. The OpenAPI description is authoritative for schemas; GET /capabilities is authoritative for values: voices, languages, dialects, formats, limits and error codes all come from there, generated from the same catalogues the request handlers validate against. Read capabilities at start-up rather than hard-coding lists from these pages.

Where to start

If you want to	Read
make your first request in five minutes	Quickstart
speak text, in real Swiss German dialect	Text-to-speech
transcribe calls, files or live audio	Speech-to-text
process long recordings or long texts	Jobs and callbacks
control pauses, numbers and pronunciation	Markup
plan capacity and understand a `429`	Limits and metering
handle errors correctly	Errors
know what is stored and what is not	Stored data

Two environments, one URL

The environment is carried by the key, not by the URL:

key	environment
`sv_…`	live	real voices and recognition, billed
`sv_test_…`	sandbox	synthetic test audio and transcripts, never billed

Every response says which one answered, in X-Suisse-Environment. Point your test suite at a sandbox key and a misconfigured environment can never reach live traffic or appear on an invoice.

Paths on this host

The documented paths start with /v1. The same endpoints also answer under /api/v1, which is the form some responses use for links (for example a job’s Location header). Both work on api.suisse-speech.ch.