Microsoft Azure has expanded its array of AI-powered voices, introducing nine new realistic options for business customers. Announced as generally available, these additions are designed to cater to a wide range of applications, from chatbots and voice assistants to e-learning and entertainment. With this latest update, Azure now supports content articulation across 91 languages and their variants, marking a significant step towards eliminating language barriers and enhancing global communication.
Diverse Applications and Styles
The new voices, accessible across all Azure regions, boast unique styles and tones to suit various needs. Among the highlights are the en-US-AvaMultilingualNeural and en-US-AndrewMultilingualNeural voices, described respectively as having a “bright, engaging female voice with a beautiful tone” and a “warm, engaging male voice“. Another voice, en-US-EmmaMultilingualNeural, is characterized as a “friendly, light-hearted, and pleasant female voice“, ideal for educational and explanatory purposes. Additionally, the En-US-RyanMultilingual voice has reached general availability, alongside the merged en-US-JennyMultilingual and en-US-JennyMultilingualV2 voices, ensuring a broad selection for users across all regions.
Ongoing Developments and Future Prospects
While the nine new voices are now generally available, Microsoft continues to test and refine additional options. The public preview includes voices such as en-IN-NeerjaNeural and hi-IN-SwaraNeural, each with three new styles available in specific regions. Furthermore, new female voices are being previewed in Central India, showcasing Microsoft’s commitment to expanding its voice technology capabilities. Alongside the introduction of realistic-sounding AI voices, Azure maintains a robust portfolio of 400 neural Text-to-Speech voices, covering over 140 languages and demonstrating the company’s dedication to fostering more inclusive and accessible communication worldwide.
Last Updated on November 7, 2024 9:19 pm CET
