Microsoft Azure Adds New Voices and Styles to Whisper or Shout in Hopeful or Unfriendly Tones
Microsoft has augmented the neural text-to-speech (TTS) options of its Azure Cognitive Services with new voices and styles of speaking. The AI-generated voices can now whisper or shout in five new voices and eight new emotions.
Azure’s TTS is designed to create realistic human voices for media, interactive voice AI, accessibility tools, and related services. The additional two female and three male voices bring the number of English-speaking voice choices to 20. All of the voices can speak with different kinds of emotion. The eight new tones include cheerful, angry, sad, excited, hopeful, friendly, unfriendly, and terrified. For the first time, there’s a naturalistic volume component to the Azure voices as well. They can be set to sound like someone who is shouting or whispering as a way of mimicking how people speak in different spaces.
“The Azure Neural TTS product team is continuously working on bringing new voice styles and emotions to the US market and beyond,” Azure AI senior product marketing manager Andy Beatman explained in a blog post. “We received feedback from customers that more voice options would help them better apply Azure Neural TTS to different user scenarios. In addition, supporting voice emotions and voice styles would help deliver the most engaging experience to end-users.”
The new choices join Azure’s newscast, customer service, and digital assistant styles added in 2020. More may be on the way as transfer learning techniques enable the neural TTS model to learn different speaking styles directly from human voice records. Microsoft also offers to produce Custom Neural Voices for clients with more specific requirements for their synthetic voices. Following the $19.7 billion acquisition of Nuance, Microsoft now has a lot more conversational and voice AI in its portfolio that will likely benefit from a wider range of options for voices and tones.
“By supporting more voice options and expanding voice styles, Azure Speech continues to address the unmet needs of the customers to build more delightful speech experience,” Microsoft Speech principal group product manager Binggong Ding said.
Follow @voicebotai Follow @erichschwartz
Microsoft Debuts Synthetic Speech Platform ‘Custom Neural Voice’
Microsoft Adds New Speech Styles and ‘Lyrical’ Emotion to Azure AI Toolkit