In a blog post today Scott Low,  Program Manager for the Microsoft Edge browser, announced that beginning with the most recent Dev and Canary channel releases, preview builds of Microsoft Edge now include support for 24 cloud-powered text to speech voices across 21 different locales.

The first place these voices will show up is in Read Aloudfeature from the current (EdgeHTML-based) version of Microsoft Edge that gives people the option to have websites read to them by the browser.  

The new voices are powered by Microsoft’s Cognitive Services and are expected to sound much better than the current robotic and unnatural voices.

The new voices come in two different styles and can be distinguished from other voices you may have installed on your computer by the fact that they have “Microsoft <voiceName> Online” in their names: 

  • Neural voices  Powered by deep neural networks, these voices are the most natural sounding voices available today. 
  • Standard voices – These voices are the standard online voices offered by Microsoft Cognitive ServicesVoices with “24kbps” in their title will sound clearer compared to other standard voices due to their improved audio bitrate.  

To try out the new array of cloud-powered voices use the Read Aloud feature. To do this, navigate to a website, select some text, right click it, and select “Read aloud selection. This will start Read Aloud and will also open the Read Aloud menu bar which lets you pick different voices and adjust reading speed by clicking on the “Voice options” button: 

Screen capture showing the "Voice options" menu in Microsoft Edge

It’s also worth noting that these voices have been exposed to developers through the JavaScript SpeechSynthesis APIThis means that any web-based text to speech application can leverage them to create more configurable and human-sounding experiences in the new version of Microsoft Edge.

What do our readers think of the new voices? Let us know below.

Comments