Exploring Microsoft Copilot’s New Audio AI: More Personal Sounding than ChatGPT

Microsoft continues to innovate with its AI tools, recently unveiling Copilot Vision, a powerful screen analysis utility. Following this launch, they’re now testing a new feature called Copilot Audio Expressions, a tool capable of generating human-like audio for your scripts. This advanced AI technology not only reads text aloud but can also creatively interpret it based on your input.

Those interested can test this tool on the official Copilot Labs website, where it features two distinct modes: Emotive and Story. Users have access to nearly a dozen voice and style variations. A detailed evaluation conducted by Windows Latest highlighted the tool’s effectiveness, particularly in its English language capabilities.

In their exploration, Windows Latest began with a fictional script depicting a bustling train station. They input the script into the tool, specifying the desired voice style. The choice was made to utilize the Emote option with the Oak voice and a narration style.

The application quickly generated an audio clip that could be played back using the built-in web player. Notably, the Emote mode allows for audio clips of up to 59 seconds in length. Users might anticipate a straightforward AI voice reciting the text; however, the output was more dynamic. The AI adeptly enhanced the script, introducing additional elements and rephrasing for increased engagement, resulting in a more captivating listening experience.

Another standout feature is the ability to download the audio file in MP3 format without requiring a user account, making playback on various devices effortless. The Emote mode showcased impressive customization options, prompting curiosity about the Story mode as well.

Crafting Engaging Narratives with Story Mode

Exploring further, Windows Latest transitioned to the Story mode, which automatically selects the voice and style without offering individual adjustments. By entering a basic prompt—”tell a story about a cat lurking in the shadows for filling prey” —the tool generated a creative 90-second narrative. This tale included a human narrator’s voice, presented with an American accent, and the cat’s unique voice delivered in a British accent, making for an entertaining audio experience.

Listeners can explore the output quality by checking both audio snippets produced by the tool.

Overall, the Copilot Audio Expressions tool stands out as an impressive AI audio generation solution, ideal for crafting engaging audio clips. If you’re interested in experiencing this innovative technology, be sure to visit the Copilot Labs site and give it a try.

Source & Images