Back to blog
Blog

Introducing AI Audio Generation

Cosmic's avatar

Cosmic

February 23, 2026

Introducing AI Audio Generation - cover image
0:00
-3:01
Listen to this article · 3:01

Cosmic AI now includes text-to-speech generation, powered by OpenAI's TTS models. Convert any text into natural-sounding audio and save it directly to your media library as an MP3, ready for CDN delivery.

Whether you're creating audio versions of blog posts, podcast intros, product walkthroughs, or accessibility-focused content, audio generation is available in the dashboard, API, and SDK.

9 Natural-Sounding Voices

Choose from a range of voices to match your content's tone:

Feminine voices:

  • Nova (default): Warm and bright, great for friendly narration
  • Shimmer: Soft and intimate, ideal for meditation or bedtime stories
  • Coral: Clear and polished, professional tone for product demos
  • Sage: Calm and steady, thoughtful pacing for education and tutorials
  • Alloy: Neutral and balanced, versatile for general-purpose use

Masculine voices:

  • Echo: Deep and authoritative, great for announcements and news
  • Onyx: Bold and commanding, strong presence for intros and branding
  • Fable: Animated and expressive, a natural storyteller for audiobooks
  • Ash: Warm and approachable, conversational for interviews

Two Quality Tiers

  • Standard (tts-1): Fast, low-latency generation. Recommended for most use cases.
  • HD (tts-1-hd): Higher quality audio with richer detail. 2x token cost.

Long Text Support

Texts over 4,096 characters are automatically split at paragraph boundaries and concatenated into a single seamless audio file. No manual chunking required.

How to Use in the Dashboard

  1. Navigate to Media in your project
  2. Click Create and select Audio
  3. Select a voice from the dropdown
  4. Paste or type the text you want to convert
  5. Click Generate

The audio file is saved to your media library in MP3 format, available instantly via CDN.

API and SDK Access

Generate audio programmatically using the API or JavaScript SDK:


Or with cURL:


Optional parameters include ( or ), , , and .

Pricing

Audio generation tokens scale with text length:

ModelCost per 1,000 characters
TTS Standard3,600 tokens
TTS HD7,200 tokens

Get Started

Audio generation is available now on all Cosmic plans. Open your project, head to Media, and try it out. For full API documentation, visit the AI API reference.

For dashboard usage details, see the AI dashboard docs.

Ready to get started?

Build your next project with Cosmic and start creating content faster.

No credit card required • 75,000+ developers