Public API

Voice AI infrastructure under your brand

Build transcription, subtitles, speaker labels, guardrails, realtime speech, and voice workflows on Whisper VoiceKit models through provider-neutral RealtimeVoiceKIT endpoints.

Usage pricing

$10 free usage, then metered billing

Every developer account starts with a $10 API credit. Paid usage is billed through Stripe from the exact SKU consumed.

$10 free usage

Rates shown are retail RealtimeVoiceKIT prices for each SKU.

Branded model catalog

Whisper VoiceKit Best

$0.252 / hour

Highest-accuracy pre-recorded speech-to-text for production transcripts.

Whisper VoiceKit Fast

$0.18 / hour

Lower-cost pre-recorded transcription with broad language coverage.

Whisper VoiceKit Realtime Pro

$0.54 / hour

Low-latency streaming speech-to-text for live products.

Whisper VoiceKit Agent

$5.40 / hour

Realtime voice-agent sessions for browser, app, and phone workflows.

API documentation

Everything you need to integrate

Authenticate with an rtvk_ bearer key and start transcribing in minutes.

Interactive reference

Explore the filtered OpenAPI schema and test API calls from the browser.

Open reference

Pre-recorded STT

Create transcripts from URLs or signed uploads and retrieve text, words, utterances, and subtitles.

View endpoints

Usage ledger

Inspect credit balance, itemized usage events, and Stripe checkout for metered billing.

View billing API

MCP server

Connect Claude, Cursor and other AI tools to your transcripts over Model Context Protocol.

Connect a client

Quickstart guide

Create an API key, upload audio and fetch the result with a few lines of code.

Read the guide

Start with a provider-neutral API

Create an rtvk_ key, use the /v2 endpoints, and keep your customer experience fully branded.