In September 2023, Microsoft"s Azure team announced it was launching a number of realistic-sounding AI voices for business customers to use. This week, Microsoft stated that nine more realistic AI voices are now generally available for those users.
In a blog post, Microsoft said these voices could be used for a wide variety of jobs and tasks, including for business chatbots, voice assistants, and e-learning jobs. They could also be used for more fun entertainment and gaming tasks.
Microsoft added:
This expansion broadens our capacity to articulate content across 91 languages and their variants. This advancement underscores our steadfast commitment to surmounting language barriers and promoting a more inclusive and accessible global communication landscape.
The nine new voices are available in all Azure regions. They include:
- en-US-AvaMultilingualNeural
- en-US-AndrewMultilingualNeural
- en-US-EmmaMultilingualNeural
- en-US-BrianMultilingualNeural
- De-DE-FlorianMultilingualNeural,
- De-DE-SeraphinaMultilingualNeural
- Fr-FR- RemyMultilingualNeural
- Fr-FR-VivienneMultilingualNeural
- zh-CN-XiaoxiaoMultiligualNeural
Each of the new voices has its own style and tone. For example, Microsoft says the en-US-AvaMultilingualNeural voice has a "bright, engaging female voice with a beautiful tone," while the en-US-AndrewMultilingualNeural voice is a " warm, engaging male voice that sounds like someone you want to know."
Some voices are better suited for specific applications. Microsoft says the en-US-EmmaMultilingualNeural version is a "friendly, light-hearted, and pleasant female voice that works well for education and explanations."
Microsoft also stated that the En-US-RyanMultilingual voice is now generally available for all Azure regions in GA. It has also merged the En-US-JennyMultilingual voice with en-US-JennyMultilingualV2 voices, and it is now available for all regions.
The company is still in a public preview with some voices. That includes the en-IN-NeerjaNeural and hi-IN-SwaraNeural voices, each with three new styles in East US, West Europe, and Southeast Asia. Also, new female voices are available in preview in Central India: en-IN-KavyaNeural, en-IN-AnanyaNeural, en-IN-AashiNeural, hi-IN-KavyaNeural, and hi-IN-AnanyaNeural.
In addition to its growing number of realistic-sounding AI voices, Microsoft Azure also has 400 neural Text-to-Speech voices, which cover over 140 languages.