Skip to main content
Chariot is a voice AI platform built for English and Indian languages. Send text and get lifelike speech (TTS) through a simple REST call, a low-latency HTTP stream, or a real-time WebSocket. Stream audio and get text back as it is spoken (speech to text, or ASR), including speech that mixes Indian languages with English.

Quickstart

Make your first speech request in a few minutes.

Choose your API

REST vs. HTTP streaming vs. WebSocket, pick the right transport.

Voices

Browse the voice catalog.

Speech to text

Transcribe live audio or upload a recording with file transcription (Beta).

API reference

Every endpoint, parameter, and response.

Credits & pricing

What each product costs.

​
What you can build

  • Voiceovers and narration, turn scripts and articles into audio.
  • Conversational agents, hear callers with streaming speech to text and answer them with low-latency speech, for IVR, support bots, and voice assistants.
  • Accessibility, read content aloud in English and Indian languages.
  • Transcription, caption live audio, transcribe phone calls, and let users dictate, in English and Indian languages.

​
How it works

1

Get an API key

Create a key in the dashboard and send it in the chariotai-api-key header. See Authentication.
2

Pick a voice

List the voice catalog and copy a voice_id.
3

Synthesize

Call POST /v1/tts for a complete file, or stream for low latency.
4

Track usage

Each response reports the credits charged; check your balance any time with GET /v1/credits.

​
Core concepts