MiniMax Audio

MiniMax Audio is a browser-based studio for speech and music generation from MiniMax. It combines text-to-speech with a voice library, voice cloning, voice design from a written description, a voice isolator, and music creation. The site lists Speech 2.8 in Turbo and HD versions, Music 3.0, and support for more than 40 languages and accents. You can type a script, pick a preset or cloned voice, adjust emotion, and generate audio for videos, podcasts, games, or ads. Developers can reach the same models through the MiniMax API.

Key Features

• Text-to-speech with the Speech 2.8 HD and Turbo models, multiple languages, and emotion controls.
• A voice library with ready-made voices for narration, characters, education, and presentations.
• Voice cloning from a short audio sample, with a separate tool for removing background sound.
• Voice design that creates a new voice from a text description.
• Music creation with Music 3.0, covering genres such as pop, electronic, R&B, jazz, country, and blues.
• A voice isolator that separates speech from background audio.
• API access for developers, plus subscriptions with credits, voice slots, and commercial-use rights.

Who is it for

• Creators making voice-overs for videos, podcasts, and social clips.
• Audiobook and explainer producers who need consistent narration.
• Game, animation, and interactive-story teams building character voices.
• Marketers and agencies producing audio ads and branded spots.
• Developers who want speech, cloning, or music models behind an API.

What can you do with MiniMax Audio

• Paste a script and generate narration in a preset voice.
• Clone a voice from a sample and reuse it across several projects.
• Describe a new voice, preview it, and then use it for a full script.
• Generate a song or background music from a short prompt.
• Isolate a voice from a noisy recording and reuse the clean audio.
• Connect the API to an app, chatbot, or automated production pipeline.

How does MiniMax Audio work

1. Sign in to the MiniMax Audio site and open the text-to-speech or music tool.
2. Choose a voice from the library, design one from text, or upload a sample to clone.
3. Enter the script or music prompt, then select a model, language, and emotion setting.
4. Generate the audio, preview it in the browser, and adjust the voice or prompt if needed.
5. Download the result or send it to a video, podcast, or game project.
6. Subscribe for more credits and voice slots, or use the API for production volume.

Pricing

• Free: $0 per month, with 10,000 credits, about 12 minutes of HD speech, about 33 songs, and 3 voice slots.
• Starter: $5 per month, or $4.50 per month billed yearly at $54, with 100,000 credits and 10 voice slots.
• Creator: $17 per month, or $13 per month billed yearly at $156, with 330,000 credits and 30 voice slots.
• Standard: $38 per month, or $25 per month billed yearly at $300, with 750,000 credits and 50 voice slots.
• Pro: $150 per month, or $80 per month billed yearly at $960, with 3 million credits and 250 voice slots.
• Paid plans list commercial-use rights for new songs and faster generation, and the site also offers API access for developers.

FAQ

Is MiniMax Audio free?

Yes. The Free plan costs $0 per month and includes 10,000 credits, about 12 minutes of HD speech, about 33 songs, and 3 voice slots. Paid plans add credits and voice slots.

Can I clone my own voice?

Yes. The voice clone tool creates a voice from a short audio sample, and the cloned voice can then be used for text-to-speech in your projects.

Does it generate music as well as speech?

Yes. The music tool uses Music 3.0 and supports several genres, including pop, electronic, R&B, jazz, country, and blues.

Which languages does it support?

The pricing page lists support for more than 40 languages and accents. The available voices and models vary by language and plan.

Is there an API?

Yes. MiniMax offers an API for its speech and music models, and the subscription page includes an API access link for developers.