canonical: https://jentic.com/apis/camb.ai/camb-ai

# CAMB.AI API

Jentic publishes the only available OpenAPI specification for CAMB.AI API, keeping it validated and agent-ready. CAMB.AI is an AI audio platform covering text-to-speech, dubbing, transcription, translation, voice generation from descriptions, sound and music generation, story narration, and source separation across 140+ languages. Ten POST endpoints expose each capability as a discrete job, including end-to-end video dubbing, translated text-to-speech, and stem separation, making it suited for media localisation, podcasting, and creative-tool agents.

## For AI agents

Generate speech, dub video, transcribe audio, translate text, and synthesize voices and sound effects across 140+ languages.

## Scope

Does not handle live real-time voice calls, video editing, or subtitle file generation - use for AI speech synthesis, dubbing, transcription, and audio generation jobs only.

## Capabilities

- Synthesize streaming text-to-speech audio in 140+ languages via /tts-stream
- Translate text and synthesize the result in a target voice via /translated-tts
- Run end-to-end video dubbing with voice cloning and lip context via /dub
- Transcribe audio recordings into text via /transcribe
- Translate text between supported languages via /translate
- Generate narrated stories and translated story variants via /story and `/translated-story/{run_id}`
- Create custom voices from a text description via /text-to-voice and synthesize sound effects or music via /text-to-sound

## Use cases

### Video Dubbing for Global Distribution

Media teams localising video for international audiences need to replace the original voice track with translated dialogue while keeping the speaker's vocal identity. The /dub endpoint runs an end-to-end pipeline that translates, voices, and aligns dubbed audio in 140+ target languages, removing the need to chain separate transcription, translation, and TTS calls.

Example prompt: Send POST /dub with a source video URL and target_language='es', then poll the returned run ID until the dubbed audio file is ready.

### Multilingual Podcast Production

Podcast networks producing content in multiple languages can chain CAMB.AI endpoints in a single workflow. /transcribe converts the original episode to text, /translate produces the localised script, and /tts-stream or /translated-tts generates the localised audio in the chosen voice, cutting localisation effort to minutes per episode.

Example prompt: Call /transcribe on the episode MP3, /translate the result to French, then /translated-tts with voice_id and the French text to render the localised audio.

### Generative Audio for Creative Tools

Creative apps that build voiceovers, audiobooks, and game audio need on-demand voice and sound generation. /text-to-voice creates a custom voice from a text description, /text-to-sound generates sound effects or music from a prompt, and /story narrates structured story content with consistent character voices.

Example prompt: Send POST /text-to-voice with description='warm male storyteller in his 50s', then use the returned voice_id in /story to narrate a five-paragraph children's tale.

### AI Agent Audio Workflows via Jentic

Through Jentic, an agent searches for 'dub a video into another language', loads the /dub input schema, and executes the call without managing the x-api-key header. Long-running jobs return run IDs that the agent can poll through subsequent Jentic-mediated calls.

Example prompt: Use the Jentic SDK to search 'dub a video', load the /dub operation schema, and execute it with a video URL and target language to start a dubbing run.

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/tts-stream` | Streaming text-to-speech synthesis |
| POST | `/dub` | End-to-end video or audio dubbing |
| POST | `/transcribe` | Transcribe audio to text |
| POST | `/translate` | Translate text between supported languages |
| POST | `/text-to-voice` | Generate a custom voice from a description |
| POST | `/text-to-sound` | Generate sound effects or music from a prompt |
| POST | `/audio-separation` | Separate vocals from instrumentals or other stems |

## Key resources

- **Speech** — Streaming and translated text-to-speech synthesis in 140+ languages
- **Dubbing** — End-to-end video dubbing combining transcription, translation, and voicing
- **Transcription** — Speech-to-text conversion for audio inputs
- **Translation** — Text translation between supported languages
- **Stories** — Story narration and translated story variants
- **Voices** — Custom voice generation from a text description
- **Audio** — Sound effect, music generation, and source separation

## Why Jentic

- **Setup:** Wiring the CAMB.AI API by hand means setting its x-api-key header and polling each dubbing, transcription, and synthesis job to completion yourself. Through Jentic you install once, import CAMB.AI from the API Directory, store the key once, and your agent calls it.
- **Permission scoping:** CAMB.AI takes its job inputs in the request body rather than as a resource in the URL path, so limit the agent to the operations it needs, such as transcribe and text-to-voice. You choose the operations it may call, so heavier jobs like dub or audio separation are not included unless you add them.
- **Credential handling:** Your CAMB.AI API key is stored once, encrypted, by your own Jentic One instance and injected into the x-api-key header at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'dub a video into another language' or 'transcribe an audio file', and Jentic returns the matching CAMB.AI operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **ElevenLabs API** — ElevenLabs focuses on high-fidelity text-to-speech and voice cloning, where CAMB.AI also covers end-to-end video dubbing and audio separation in one platform.
- **Deepgram API** — Deepgram specialises in low-latency speech-to-text, where CAMB.AI offers transcription alongside generative audio and dubbing.
- **AssemblyAI API** — AssemblyAI provides advanced transcription with summarisation and speaker analytics that pairs well with CAMB.AI's dubbing and voice generation.

## FAQ

### Why is there no official OpenAPI spec for CAMB.AI API?

CAMB.AI does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call CAMB.AI API via structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer.

### What authentication does the CAMB.AI API use?

It uses an API key passed in the x-api-key HTTP header. Through Jentic, the key is stored in your Jentic One instance and injected into the x-api-key header at execution time, so the raw key never enters the agent's prompt context.

### Can I dub a video end-to-end with one call?

Yes. POST /dub accepts a source media reference and target language and runs transcription, translation, and voicing as a single asynchronous job. The response includes a run ID that you poll to retrieve the dubbed media when processing completes.

### How do I transcribe and translate a podcast through Jentic?

Use the Jentic SDK to search 'transcribe and translate audio'. Jentic returns the /transcribe and /translate operations with their input schemas. Execute /transcribe with the audio URL, then pass the returned text into /translate with the target language.

### What languages does the API support?

CAMB.AI advertises support for 140+ languages across speech synthesis, dubbing, and translation. Pass the target language code in the request body for /translated-tts, /dub, or /translate; check the CAMB.AI docs for the full ISO code list.

### Are jobs synchronous or asynchronous?

Long-running operations such as /dub, /story, and `/translated-story/{run_id}` are asynchronous and return a run ID. Poll the run-specific endpoint with that ID until the job reports completion before downloading the resulting media.

### Can I limit what my agent is allowed to do with the CAMB.AI API?

Yes. Because you run Jentic One yourself, your own rules decide which CAMB.AI operations and credentials the agent may use. Since CAMB.AI passes its job inputs in the request body rather than as a resource in the URL path, you scope access by choosing the specific operations the agent can call, such as /transcribe and /text-to-voice. Heavier jobs like /dub or /audio-separation stay off limits unless you add them, and your API key is injected at execution time so the agent never sees it.