New Copilot Audio Expressions experimental tool

Generate speech in seconds with Copilot Audio Expressions​

Copilot Audio Expressions is a Copilot Labs experiment that transforms written text into natural, spoken narration—complete with emotion, personality, and style. You can adjust how the voice sounds to match your prompt, whether it calls for a gentle tone for a bedtime story, an upbeat message for a motivational pep talk, or theatrical flair for an epic adventure. Easily download audio samples to use for content creation.

It also includes Story Mode, which weaves together multiple vocal styles to generate engaging, accessible storytelling experiences—perfect for parents, educators, etc. Audio Expressions offers a glimpse into how AI can make audio content more immersive, expressive, and fun for everyone.

Available Features​

In Copilot Audio Expressions, you can experiment with voice settings to explore different styles of narration:

• Using Emotive Mode, provide a script outline (prompt) and have Copilot perform it in a variety of styles.
  • Adjust the narrator’s tone and pace—from a calming whisper to an excited cheer—to match the mood of each scene.
  • Assign a distinct voice so your dialogue feels dynamic and alive.
• Try Story Mode to hear your own stories brought to life with expressive narration.
• Download audio samples to use in your personal projects

Frequently Asked Questions​

Copilot Audio Expressions is free to use. It’s available globally and works on any modern browser. Currently, this experimental tool supports English only. For the best experience, we recommend using a desktop computer, though it’s also accessible via mobile browsers.

What’s the difference between Emotive Mode and Story Mode?

• Emotive Mode – You provide your own script outline as a prompt, choose a specific voice character and style, and Copilot enhances it for optimal listening.
• Story Mode – You share a story idea or short prompt, and Copilot crafts a custom script, selects the most fitting voice character, and brings your story to life.

How many audio expressions can I generate and download?
There's no set limit, however this may change based on the demand. Each expression varies in length - for a given prompt, the tool typically generates about 30 seconds (Emotive Mode) to about 90 seconds (Story Mode) of speech.

What file format will my audio download in?
All audio expressions are provided in high-quality MP3 format, ready for sharing, embedding, or playback across devices.

Is there any guidance on what I can use Copilot Audio Expressions for?
Copilot Audio Expressions is intended for personal use that aligns with the Copilot Code of Conduct.

Keep in mind — you should only input content you own or have the rights to use. Do not input copy or request audio that infringes on the rights of others (including copyright, trademark, or publicity rights), violates the privacy of others, or otherwise violates the Microsoft Services Agreement or Copilot Terms.

Your account or Copilot access may be restricted or suspended if we detect an attempt to input or generate illegal content or content that violates the Copilot Terms Code of Conduct.