[Integrations](https://spiral.ai/integrations) › [Codex](https://spiral.ai/integrations/codex) › [MiniMax Voice Cloning](https://spiral.ai/integrations/codex/minimax-voice-cloning)

# How to Use MiniMax Voice Cloning in Codex

MiniMax Voice Cloning turns a voice recording into a reusable voice for speech generation. Use its optional preview to audition narration before producing longer audio with text-to-speech.

|  |  |
| --- | --- |
| Model | MiniMax Voice Cloning |
| Available through | fal |
| Category | Audio |
| Price | $1.95 per voice, plus $0.39 per 1,000 characters of preview text. |
| Returns | audio, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with MiniMax Voice Cloning in Codex

- Use MiniMax Voice Cloning with recordings/narrator.wav and a speech-02-hd preview. Extract a passage under 700 characters from docs/tutorial.md that includes the product name and two technical terms.
- Use MiniMax Voice Cloning with recordings/host.mp3 to audition the introduction in scripts/podcast-intro.txt. Keep the speech-02-hd preview under 1,000 characters and save the returned voice reference beside the audition.
- Use MiniMax Voice Cloning with recordings/assistant.wav and speech-02-turbo. Build a preview under 500 characters from the greeting and first help response in the project's dialogue fixtures.
- Use MiniMax Voice Cloning to audition recordings/instructor.m4a. Select an English lesson excerpt under 800 characters from content/lessons and generate a speech-02-hd preview for review.
- Use MiniMax Voice Cloning with recordings/spanish-narrator.wav and speech-02-hd. Take a Spanish passage under 700 characters from locales/es/tutorial.json and save it alongside the preview for a fluent reviewer.

## How to set up MiniMax Voice Cloning in Codex

Pick the way that suits you. Each takes about two minutes.

### Use Spiral's MCP server

1. **Add the hosted server**

```bash
codex mcp add spiral --url https://api.spiral.ai/mcp
```

2. **Sign in**: Codex opens the browser: sign in to Spiral and click Authorize. No API key. `codex mcp list` then shows Spiral.

```bash
codex mcp login spiral
```

### Use the Spiral skill

1. **Send Codex this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run MiniMax Voice Cloning**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "fal-ai/minimax/voice-clone",
    "input": {
      "audio_url": "https://storage.googleapis.com/falserverless/model_tests/zonos/demo_voice_zonos.wav"
    }
  }'
```

Then paste one of the prompts above: Codex finds MiniMax Voice Cloning on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/codex

## MiniMax Voice Cloning tools and prices

- [MiniMax Voice Cloning](https://spiral.ai/tools/fal/fal-ai/minimax/voice-clone) (fal): $1.95 per voice, plus $0.39 per 1,000 characters of preview text.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| noise_reduction |  | false | Enable noise reduction for the cloned voice |
| need_volume_normalization |  | false | Enable volume normalization for the cloned voice |
| accuracy | 0 to 1 | — | Text validation accuracy threshold (0-1) |
| text |  | Hello, this is a preview of your cloned voice! I hope you like it! | Text to generate a TTS preview with the cloned voice (optional) |
| model | speech-02-hd, speech-02-turbo, speech-01-hd, speech-01-turbo | speech-02-hd | TTS model to use for preview. Options: speech-02-hd, speech-02-turbo, speech-01-hd, speech-01-turbo |

## What can you do with MiniMax Voice Cloning in Codex?

Give a project's narrator or assistant a voice derived from a recording. Your coding agent can prepare an audition passage from project content and keep the approved voice reference for later speech generation.

### Audition an audiobook narrator

Create a narrator voice from a recording, then preview a short passage with dialogue and descriptive prose using Speech 02 HD.

### Prepare a course narration voice

Clone an instructor's voice and preview a lesson excerpt containing the subject's technical vocabulary before producing the course narration.

### Test product tutorial narration

Audition a presenter voice on a short explanation containing product names, numbers, and interface labels.

### Develop a podcast introduction

Create a host voice from a recording and preview an opening script before using the approved voice in later speech generation.

### Review multilingual narration

Preview a cloned narrator with a translated passage using Speech 02. Have a fluent reviewer check pronunciation and delivery in the target language.

### Audition a conversational assistant voice

Preview a cloned voice reading a greeting and a short answer with Speech 02 Turbo before selecting it for conversational speech.

## Overview

- [fal](https://spiral.ai/providers/fal)
- [All Codex guides](https://spiral.ai/integrations/codex)

## MiniMax Voice Cloning in other agents

- [Amp](https://spiral.ai/integrations/amp/minimax-voice-cloning)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/minimax-voice-cloning)
- [Claude](https://spiral.ai/integrations/claude-desktop/minimax-voice-cloning)
- [Claude Code](https://spiral.ai/integrations/claude-code/minimax-voice-cloning)
- [Cline](https://spiral.ai/integrations/cline/minimax-voice-cloning)
- [Cursor](https://spiral.ai/integrations/cursor/minimax-voice-cloning)
- [Devin](https://spiral.ai/integrations/devin/minimax-voice-cloning)
- [dots](https://spiral.ai/integrations/dots/minimax-voice-cloning)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/minimax-voice-cloning)
- [Gemini](https://spiral.ai/integrations/gemini/minimax-voice-cloning)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/minimax-voice-cloning)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/minimax-voice-cloning)
- [goose](https://spiral.ai/integrations/goose/minimax-voice-cloning)
- [Grok](https://spiral.ai/integrations/grok/minimax-voice-cloning)
- [Grok Build](https://spiral.ai/integrations/grok-build/minimax-voice-cloning)
- [Hermes](https://spiral.ai/integrations/hermes/minimax-voice-cloning)
- [Instinct](https://spiral.ai/integrations/instinct/minimax-voice-cloning)
- [Manus](https://spiral.ai/integrations/manus/minimax-voice-cloning)
- [Muse](https://spiral.ai/integrations/muse/minimax-voice-cloning)
- [Muse Code](https://spiral.ai/integrations/muse-code/minimax-voice-cloning)
- [n8n](https://spiral.ai/integrations/n8n/minimax-voice-cloning)
- [Notion](https://spiral.ai/integrations/notion/minimax-voice-cloning)
- [OpenClaw](https://spiral.ai/integrations/openclaw/minimax-voice-cloning)
- [opencode](https://spiral.ai/integrations/opencode/minimax-voice-cloning)
- [Perplexity](https://spiral.ai/integrations/perplexity/minimax-voice-cloning)
- [Raycast](https://spiral.ai/integrations/raycast/minimax-voice-cloning)
- [VS Code](https://spiral.ai/integrations/vscode/minimax-voice-cloning)
- [Windsurf](https://spiral.ai/integrations/windsurf/minimax-voice-cloning)
- [Zapier](https://spiral.ai/integrations/zapier/minimax-voice-cloning)

## More Codex guides

- [MiniMax Speech-02 Turbo](https://spiral.ai/integrations/codex/minimax-speech-02-turbo)
- [MiniMax Voice Design](https://spiral.ai/integrations/codex/minimax-voice-design)
- [MiniMax Speech-02 HD](https://spiral.ai/integrations/codex/minimax-speech-02-hd)
- [Mirelo SFX 2.0 Text to Audio](https://spiral.ai/integrations/codex/mirelo-sfx-2-0-text-to-audio)
- [MiniMax Speech 2.8 Turbo](https://spiral.ai/integrations/codex/minimax-speech-2-8-turbo)
- [Mirelo SFX 2.0 Video to Audio](https://spiral.ai/integrations/codex/mirelo-sfx-2-0-video-to-audio)

## Questions about MiniMax Voice Cloning in Codex

**Can Codex use MiniMax Voice Cloning?**

Yes. Add Spiral to Codex as an MCP server, and Codex can find and run MiniMax Voice Cloning like any built-in tool.

**How much does MiniMax Voice Cloning cost in Codex?**

$1.95 per voice, plus $0.39 per 1,000 characters of preview text. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.

**Do I need a fal account to use MiniMax Voice Cloning in Codex?**

No. Spiral runs MiniMax Voice Cloning on its own fal account. You sign in to Spiral once and never paste a key.

**How long does it take to set up MiniMax Voice Cloning in Codex?**

About two minutes: add the hosted server, then sign in.

**Does Codex wait while MiniMax Voice Cloning runs?**

MiniMax Voice Cloning runs in the background. Codex starts the run, checks on it and gives you the audio, json when it's ready.

**What does MiniMax Voice Cloning create?**

It creates a reusable voice from a recording and can generate a spoken preview. Longer narration uses that cloned voice in a separate text-to-speech generation.

**How much reference audio does MiniMax Voice Cloning need?**

Use an MP3, M4A, or WAV recording lasting 10 seconds to 5 minutes. The recording must be no larger than 20 MB.

**How long can the spoken preview be?**

Preview text can contain up to 1,000 characters. Use a representative passage to judge pronunciation and delivery before producing longer narration.

**Should I use Speech 02 HD or Speech 02 Turbo for the preview?**

Choose HD to audition voiceovers or audiobook narration. Turbo prioritizes low latency for interactive speech, so use it when evaluating a voice for conversational responses.

**Can I preview the cloned voice in different languages?**

With a Speech 02 preview, MiniMax documents support for 32 languages. Test the actual language and terminology you plan to use before approving the voice.

