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

# AudioPod AI API

The AudioPod AI API generates and processes audio: it synthesizes speech, clones and converts voices, generates music and instrumentals, transcribes and summarizes recordings, and separates speakers and stems. Jobs run asynchronously with status polling, and higher-level project workflows assemble multi-chapter audiobooks and multi-speaker podcasts from a manuscript or sources. It also converts media formats, downloads source audio, and meters usage through a credit wallet.

## For AI agents

Generate speech, clone and convert voices, transcribe audio, generate music, and separate speakers and stems. Authenticates with an X-API-Key header (recommended for agents) or an OAuth2 bearer token.

## Scope

Does not run a public media player, handle music royalties, or replace your audio editing software. Use for generating, transcribing, and processing audio only.

## Capabilities

- Generate speech from text and stream it, with per-voice settings
- Clone a voice from samples and convert one voice to another
- Transcribe audio to text, then summarize or edit the transcript
- Separate speakers, extract stems, and denoise audio
- Generate music, instrumentals, and covers from a prompt or lyrics
- Produce multi-chapter audiobooks and multi-speaker podcasts

## Use cases

### Agent Voiceover and Narration

An AI agent connected through Jentic turns a script into spoken audio with a chosen or cloned voice, streaming or polling the job until it is ready. Because voice settings and streaming are first-class, the agent can produce narration for a video or notification without a human in the editing loop.

Example prompt: Generate speech for this paragraph with a chosen voice and return the audio once the job finishes

### Transcription and Summary Pipeline

A workflow submits a recording for transcription, waits for the job, then summarizes and edits the resulting transcript, so an agent can turn a meeting or interview into text and a short summary. The edit and summarize steps keep the output usable without a separate tool.

Example prompt: Transcribe this recording, then return a short summary of the transcript

### Music and Stem Workflows

A creator generates a music track from a text prompt or separates an existing song into stems, running each as a job with status polling. The stem extraction and music generation endpoints let an agent build or deconstruct audio on demand.

Example prompt: Generate a 30-second instrumental from a prompt, then separate it into stems

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/api/v1/voice/voices/{voice_identifier}/generate` | Generate voice audio |
| POST | `/api/v1/voice/voice-clone` | Create a voice cloning job |
| POST | `/api/v1/transcription/transcribe` | Create a transcription job |
| POST | `/api/v1/music/text2music` | Generate music from text |
| POST | `/api/v1/speaker/separate` | Separate voices from audio |
| POST | `/api/v1/audio/speech` | Create speech |

## Key resources

- **Voices** — Voice profiles, cloning, conversion, and text-to-speech generation
- **Transcription** — Speech-to-text jobs with summary, editing, and versioned transcripts
- **Music and stems** — Prompt-based music generation and stem extraction and separation jobs
- **Projects** — Audiobook and podcast production workflows assembled from manuscripts or sources

## 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:** 58 / 100
- **Maturity:** Foundational
- **Dimensions:**
  - Foundational Compliance: 62 / 100
  - Developer Experience & Jentic Compatibility: 67 / 100
  - AI-Readiness & Agent Experience: 74 / 100
  - Agent Usability: 43 / 100
  - Security: 46 / 100
  - AI Discoverability: 66 / 100
- **View full report:** https://jentic.com/apis/audiopod.ai/audiopod/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 AudioPod AI API by hand means choosing between an X-API-Key header and an OAuth2 bearer token, submitting long-running jobs and polling each for status, and metering spend against a credit wallet. Through Jentic you install once, import AudioPod from the API Directory, store the key once, and your agent calls it.
- **Permission scoping:** You choose which operations your agent may call, so you can allow speech generation and transcription while leaving voice deletion, wallet top-ups, and billing out. AudioPod puts job ids in the URL path, so a rule can also bound the agent to the jobs it creates and reads back, and every call it makes is logged.
- **Credential handling:** Your AudioPod 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 speech from text' or 'transcribe an audio file', and Jentic returns the matching AudioPod operation with its input schema so the agent calls the right endpoint without reading the reference docs.

## Related APIs

- **ElevenLabs API** — Text-to-speech and voice cloning with expressive AI voices
- **Murf API** — AI voiceover and text-to-speech for studio-style narration
- **Deepgram API** — Fast, accurate speech-to-text and audio intelligence
- **AssemblyAI API** — Speech-to-text with summarization and audio understanding models

## FAQ

### Is there an AudioPod MCP server?

You don't need an MCP server to give your agent the AudioPod AI API. Jentic connects it directly from the API Directory: import it, store your key once, and your agent can generate speech, transcribe audio, or run music jobs straight away, with no extra server to run and no tool definitions to load into the agent's context.

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

Yes. Write a rule that allows only the generation and transcription operations the agent needs, such as text-to-speech and speech-to-text, and leaves voice deletion, billing, and wallet top-ups out, and every call it makes is logged. Because job ids are in the URL path, you can also bound the agent to reading back the jobs it created.

### What authentication does the AudioPod AI API use?

The AudioPod AI API authenticates with an API key in an X-API-Key header, recommended for server and agent use, or an OAuth2 password bearer token (a JWT) for session access, per its OpenAPI spec. A few endpoints are public. Through Jentic the key is stored encrypted by your own instance and added at call time, so it never appears in the agent's prompt or logs.

### Can I clone a voice with the AudioPod AI API?

Yes. You submit audio samples to create a cloning job, poll it for status, and then generate speech with the cloned voice, with a conversion endpoint to turn one voice into another. Reporting and retry operations let an agent recover a failed job rather than starting over.

### What are the rate limits for the AudioPod AI API?

The API describes published rate limits but the version indexed here does not state specific numbers in the spec. For current limits and plan quotas, see the AudioPod documentation at https://audiopod.ai.

### How do I generate speech with the AudioPod AI API through Jentic?

Import the AudioPod AI API from the Jentic API Directory, then have your agent issue a request such as 'generate speech for this text with a chosen voice'. Jentic matches the intent to the voice-generation operation and returns its input schema so the agent builds the correct request, with your stored key injected at call time. To run it on your own infrastructure, install Jentic One from its GitHub repo.