Microsoft’s Copilot AI text-to-speech gets new, cleaner ‘scripted mode’

Bookmark (0)
Please login to bookmark Close

Microsoft’s AI Manager Mustafa Suleyman recently unveiled in a social media post a new feature called “Scripted Mode” in Copilot Labs for turning written scripts into text-to-speech. Copilot Labs is an experimental platform where you can try out Microsoft’s newest AI features that are still in development.

Specifically, the feature is part of Copilot Audio Expressions, which is the actual tool that turns text into spoken audio. Previously, Copilot Audio Expressions only had two modes: Story Mode (which weaves together multiple vocal styles and characters for a storytelling experience) and Emotive Mode (which uses a single, distinct voice that matches a particular mood with some improvisation).

With Scripted Mode, Copilot Audio Expressions can now take text and read it aloud with a fast and direct take, making sure to read the content verbatim without any riffing or creative changes. You can still select which voice and style to use for the reading.

Copilot Audio Expressions is still only available in English, but Microsoft is exploring ways to support more languages.

Microsoft’s AI Manager Mustafa Suleyman recently unveiled in a social media post a new feature called “Scripted Mode” in Copilot Labs for turning written scripts into text-to-speech. Copilot Labs is an experimental platform where you can try out Microsoft’s newest AI features that are still in development.

You asked, we shipped! Scripted mode just dropped for audio generation in Copilot Labs (c/o our new MAI-Voice-1 model).Scripted mode: reads your input verbatimEmotive: riffs a bit for max dramaStory: performs multiple voices/charactersTry out all 3 ➡️ https://t.co/9hL81LTFwF pic.twitter.com/rOVZKGbDjX— Mustafa Suleyman (@mustafasuleyman) September 10, 2025

Specifically, the feature is part of Copilot Audio Expressions, which is the actual tool that turns text into spoken audio. Previously, Copilot Audio Expressions only had two modes: Story Mode (which weaves together multiple vocal styles and characters for a storytelling experience) and Emotive Mode (which uses a single, distinct voice that matches a particular mood with some improvisation).

With Scripted Mode, Copilot Audio Expressions can now take text and read it aloud with a fast and direct take, making sure to read the content verbatim without any riffing or creative changes. You can still select which voice and style to use for the reading.

Copilot Audio Expressions is still only available in English, but Microsoft is exploring ways to support more languages. Read More