# Music, sound and speech. One balance.

Source: https://corespeed.io/integrations/audio-generation

Audio generation gives Claude Code, Codex, Cursor or any MCP agent dozens of hosted music, sound effect and speech models through CoreSpeed, with no account or API key of their own. Your agent reads a model's rate, then makes a score, a sound effect or a voice-over from text. Each file lands in your organization's Drive.

## Set up an agent

Give your agent this line: `set up https://corespeed.io/SKILL.md (read it raw, e.g. with curl)`

It fetches the skill at https://corespeed.io/SKILL.md, adds the CoreSpeed MCP server, signs you in (or up), and keeps the skill for next time. Nothing to connect: Audio generation is built in. Each call is priced in credits, from your one balance. Create an account: https://app.corespeed.io/sign-up

## Describe the sound. Get the file

Your agent finds a music, sound or speech model, reads its rate, and runs it. The audio comes back as a link and as a file in your Drive.

You ask your agent: "Make a 30-second upbeat instrumental for our launch video, with light percussion and no vocals. Show me the model's rate before you run it."

It calls CoreSpeed:

- `media__list_models`: ok, 388 ms
- `media__get_model`: ok, 176 ms
- `media__generate`: ok, 21400 ms

It reads from: Audio generation.

### Audio: A 30-second launch track.

The rate was shown before the run. The track is in your Drive for 7 days.

**Done**

- One upbeat instrumental, light percussion, no vocals.

**Next**

- Lay it under the launch video, or ask for a second take.

_Illustrative example · Fictional track and timings_

How media generation works: https://corespeed.io/docs/media

## A score, a cue, a voice

Requests you can make right after setup. The data is fictional.

### Score a short video

30 seconds, instrumental · Music model

- Mood: Upbeat, light percussion
- Length: 30 seconds
- One track in your Drive

### Make a sound effect

A whoosh for the logo · Sound effect model

- Asked for: A soft whoosh, under one second
- Result: One short audio file
- Each model has its own rate

### Pick a voice for the voice-over

A narrator from a voice library · Speech + Fish Audio

- Voice: A licensed library voice
- Or: A new voice from a description
- Fish Audio's voices, the same tool

Full example: https://corespeed.io/integrations/fish-audio (Markdown: https://corespeed.io/integrations/fish-audio.md)

## Keep your agent. Nothing to connect

One CoreSpeed account, across compatible agents.

1. **Set up CoreSpeed**: Give your agent the setup line. It adds CoreSpeed and signs you in.
2. **Check the price**: Every tool has a rate in credits. Your agent reads it before each call.
3. **Start with one task**: Ask for one short track or sound, and ask your agent for the rate first.

Setup guides: Claude Code https://corespeed.io/docs/mcp#claude-code · Codex https://corespeed.io/docs/mcp#codex · Cursor https://corespeed.io/docs/mcp#cursor

## Questions

### What kinds of audio can my agent make?

Music, sound effects and speech, plus tools that clean up, split or change a recording. The catalog page lists every model with its rate. Every audio model: https://corespeed.io/catalog/audio-generation

### How is text to speech priced?

Most speech models charge per 1,000 characters of text, and each model has its own rate. Your agent reads it before it runs.

### Am I charged for a failed generation?

No. Audio is billed once, after it succeeds. Failed, cancelled and expired generations are never charged.

### Do I need an account or API key for Audio generation?

No. Audio generation is built into CoreSpeed. Your agent sees its tools after setup, with nothing to connect.

### What does Audio generation cost?

Each call has a rate in credits, shown before it runs, from your one balance. A failed call costs nothing. Every Audio generation tool and its price: https://corespeed.io/catalog/audio-generation

### Do I have to switch agents?

No. CoreSpeed adds tools to the agent you already use: Claude Code, Codex, Cursor, or another MCP client. Each one signs in once.

Every tool and price: https://corespeed.io/catalog/audio-generation

## Related pages

- [Fish Audio for AI agents](https://corespeed.io/integrations/fish-audio): Fish Audio's voices in your agent, with no Fish Audio account: search the voice library, design a new voice, and speak a script or a whole scene.
- [Tools](https://corespeed.io/tools): Models, the live web and nine social platforms, on one balance. No account for any of them.
- [Transcription for AI agents](https://corespeed.io/integrations/transcription): Turn a call, an interview or a video into text from your agent, with speaker labels and word timings. No account or API key to set up.
- [Video generation for AI agents](https://corespeed.io/integrations/video-generation): Give your agent hundreds of text to video and image to video models, with no API keys. It reads the rate first, and long renders finish as jobs.

## Sources

- Media docs: https://corespeed.io/docs/media
- Audio generation catalog: https://corespeed.io/catalog/audio-generation
- One prompt, 11 videos at once: https://corespeed.io/blog/one-prompt-eleven-videos

Reviewed 2026-11-14.

## Links

- Product overview: https://corespeed.io/ (Markdown: https://corespeed.io/index.md)
- All integrations: https://corespeed.io/integrations (Markdown: https://corespeed.io/integrations.md)
- Documentation: https://corespeed.io/docs
- Pricing: https://corespeed.io/pricing (Markdown: https://corespeed.io/pricing.md)
