Camb.AI
Camb.AI is a generative voice AI platform for translation, dubbing, and speech. Its research models - MARS (text-to-speech and voice cloning) and BOLI (neural translation) - power an API covering text-to-speech, end-to-end video and audio dubbing, text translation across 140+ languages, voice discovery and custom voice cloning, and speech-to-text transcription. The REST API lives at https://client.camb.ai/apis, is authenticated with an x-api-key header, and follows an asynchronous create-and-poll pattern (POST to start a task, GET the task by id until it succeeds, then fetch the result). Camb.AI also exposes public WebSocket surfaces for real-time streaming TTS, live transcription, and realtime speech-to-speech translation.
APIs
Camb.AI Text-to-Speech API
Convert text into natural speech with the MARS voice models. Includes a streaming endpoint (POST /tts-stream) that returns a binary audio stream, a deprecated task-based create ...
Camb.AI Dubbing API
End-to-end video and audio dubbing that preserves emotional delivery. Submit a media URL with a source language and one or more target languages (POST /dub), poll the task (GET ...
Camb.AI Translation API
Neural text translation powered by the BOLI model across 140+ language pairs, with controls for formality, gender, and specialized vocabulary. Start a translation task with an a...
Camb.AI Voices API
Discover and manage the voices used across TTS and dubbing. List public, shared, and custom voices (GET /list-voices), clone a custom voice from reference audio (POST /create-cu...
Camb.AI Transcription API
Speech-to-text transcription with speaker identification and word-level timestamps. Submit a media file or media URL with a language tag (POST /transcribe) and poll the task (GE...
Collections
Camb.AI API
OPENPricing Plans
Rate Limits
FinOps
Camb Ai Finops
FINOPSEvent Specifications
Camb.AI Realtime WebSocket API
Camb.AI exposes public WebSocket channels for real-time voice AI in addition to its REST API. Three streaming surfaces are documented: live text-to-speech (stream text in, recei...
ASYNCAPI