canonical: https://jentic.com/apis/cvoice.ai/cvoice

# cvoice.ai API

The cvoice.ai API generates AI voice audio from text using a catalog of more than 20,000 character voices. A single generation call turns text into audio for a chosen voice ID, and two public datasets list the available characters and their voice IDs so callers can pick a voice before generating. An endpoint also returns the information tied to your API key.

## For AI agents

Generate AI voice audio from text using any of 20,000+ character voices, and look up voice IDs from the public character and voice datasets.

## Scope

Does not handle speech-to-text, music generation, or audio hosting. Use for text-to-speech voice generation only.

## Capabilities

- Generate AI voice audio from text with a chosen voice ID
- Search more than 20,000 characters and their metadata in the public characters dataset
- Browse voice IDs grouped by character in the public voices dataset
- Read the information tied to your API key

## Use cases

### Agent-driven voiceover

An AI agent producing narration can pick a voice from the character catalog and generate audio for a line of text in a single call. Because the cvoice.ai API exposes the voice catalog as a public dataset, the agent resolves a character to a voice ID first and then generates, rather than guessing an identifier.

Example prompt: Search the characters dataset for a voice, then generate audio from a line of text using the matching voice ID

### Voice discovery

Choosing the right voice means browsing what is available. The cvoice.ai API publishes a characters dataset with metadata and a voices dataset grouping voice IDs by character, so a tool can present options and let a user or agent select before any audio is generated. Both datasets are public and need no key to read.

Example prompt: Load the characters dataset and the voices dataset and present the matching voice IDs for a chosen character

### Script-to-audio pipeline

An app that converts written scripts into audio can call the generation endpoint for each line with a consistent voice ID. Keeping the same voice ID across calls produces a continuous narration, and the key-info endpoint lets the pipeline confirm the key is active before a long run.

Example prompt: Generate audio for each line of a script using one voice ID to keep the narration consistent

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/api/tts` | Generate AI voice audio from text for a chosen voice ID |
| GET | `/api/me` | Read the information tied to your API key |
| GET | `/dataset_search.json` | Public characters dataset with metadata |
| GET | `/dataset_voices.json` | Public voices dataset grouped by character |

## Key resources

- **Text to speech** — Generate AI voice audio from text for a chosen voice ID
- **Datasets** — Public characters and voices datasets listing available voice IDs and metadata
- **Account** — Read the information tied to your API key

## AI readiness

This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.

- **Score:** 71 / 100
- **Maturity:** AI-Aware
- **Dimensions:**
  - Foundational Compliance: 100 / 100
  - Developer Experience & Jentic Compatibility: 67 / 100
  - AI-Readiness & Agent Experience: 59 / 100
  - Agent Usability: 94 / 100
  - Security: 50 / 100
  - AI Discoverability: 81 / 100
- **View full report:** https://jentic.com/apis/cvoice.ai/cvoice/scorecard
- **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/
- **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview

### Score it yourself

Every API in the directory is allowlisted, so you can re-score it with no key required.

- **Score your own API:** https://jentic.com/scorecard.md
- **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md

```sh
npx @jentic/api-scorecard-cli score <openapi-url>
```

## Why Jentic

- **Setup:** Wiring the cvoice.ai API by hand means handling its X-API-Key header auth, resolving a voice ID from the public datasets, and distinguishing keyed generation from the open dataset reads yourself. Through Jentic you install once, import cvoice from the API Directory, store the key once, and your agent calls it.
- **Permission scoping:** The voice ID and text travel in the request body, not the URL path, so a rule bounds which operations your agent may call rather than which voice it uses. You choose the operations it may call, such as voice generation, so reading your key information is not included unless you add it.
- **Credential handling:** Your cvoice key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'generate voice audio from text' or 'find a voice ID', and Jentic returns the matching cvoice operation with its input schema so the agent calls the right endpoint without reading the reference docs.

## Related APIs

- **ElevenLabs** — ElevenLabs generates expressive synthetic speech and voice clones, where cvoice offers a large catalog of character voices.
- **Murf** — Murf produces studio-style voiceovers from text, where cvoice focuses on a broad set of character voices.
- **OpenAI** — OpenAI generates the script text that cvoice then turns into audio.

## FAQ

### What authentication does the cvoice.ai API use?

The cvoice.ai API authenticates with an API key sent in the X-API-Key header, prefixed cvai_, as declared in its OpenAPI spec. The two public voice datasets need no key to read. Through Jentic the key is stored once and injected at call time, so it never appears in your agent's prompt or logs.

### How do I find a voice ID for the cvoice.ai API?

Read the public datasets: the characters dataset lists more than 20,000 characters with metadata, and the voices dataset groups voice IDs by character. An agent resolves a character to a voice ID from these before calling the generation endpoint.

### What are the rate limits for the cvoice.ai API?

The spec describes cvoice.ai as a free API with no usage limits, and does not state numeric rate limits. Check the vendor documentation at https://cvoice.ai for current terms before running large batches.

### How do I generate voice audio through Jentic with the cvoice.ai API?

Search Jentic for 'generate voice audio from text', which returns the POST `/api/tts` operation with its input schema, then call it with the text and a voice ID. To run it on your own infrastructure, install Jentic One from its GitHub repo.

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

Yes. Write a rule that allows only the voice-generation operation, so the agent can turn text into audio but cannot read your key information, and every call it makes is logged. The voice ID and text travel in the request body, so the rule bounds which operations run rather than which voice is used.