Camlin Speech / Documentation

Build with Camlin Speech

Australian-English speech recognition and voice synthesis, designed for contact-centre platforms, browser SDKs, accessibility workflows, and customer-hosted deployments. Stable APIs, clear examples, and production-ready fallbacks.

Speech-to-Text

STT user guide

WAV upload, JSON transcripts, word timing, confidence, recognition profiles, and per-tenant vocabulary biasing.

Open STT docs

Text-to-Speech

TTS user guide

Australian English voices, WAV audio output, per-request tuning controls, and a lip-sync frame stream for avatars.

Open TTS docs

What's in the box

One service exposing speech recognition and synthesis under /api/v1/asr/* and /api/v1/tts/*.
REST request and response shapes designed for simple browser, server, and contact-centre integration.
Production routing with fallback behavior kept behind the API contract, so clients do not need provider-specific code.
Per-tenant lexicons for names, agencies, suburbs, products, and other domain vocabulary.
Live demos for upload transcription, real-time speech, and voice synthesis.

Where to read next

If you’re wiring the API into an application: STT · recognize endpoint and TTS · synthesize endpoint are the two pages to start on.

The same content is shipped as plain markdown alongside the source under docs/stt.md and docs/tts.md for engineers reading from a checkout.