At the end of last month, Microsoft released nine more realistic AI text-to-speech voices:
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-XiaoxiaoMultilingualNeural
Characteristics
Whether for creating voice-based chatbots, voice assistants, or conversational agents, these new voices will ensure interactions are more realistic, lifelike, and engaging. Voices optimized for conversation sound more natural and engaging when reading dialogues and informal texts. Including interjections like laughter and filler pauses, adding a human touch to virtual conversations.
Check out the before and after comparison.
Old version (more awkward)
New version (very natural)
Comparison
en-US-AvaMultilingualNeural (female): A bright, engaging female voice with a beautiful tone. en-US-AndrewMultilingualNeural (male): A warm, engaging male voice that sounds like someone you'd want to get to know. en-US-EmmaMultilingualNeural (female): A friendly, easygoing female voice, perfect for education and explanation. en-US-BrianMultilingualNeural (male): A young, cheerful, and versatile voice, suitable for a wide range of scenarios. De-DE-FlorianMultilingualNeural (male): German male voice De-DE-SeraphinaMultilingualNeural (female): German female voice Fr-FR-RemyMultilingualNeural (male): French male voice Fr-FR-VivienneMultilingualNeural (female): French female voice zh-CN-XiaoxiaoMultilingualNeural (female): Chinese female voice
Use
To use, you can visit ai.azure.com to listen to demos from the voice library or activate the voice feature in the ChatGPT Playground. This enhancement facilitates seamless audio testing and provides a more refined and realistic conversational experience for integration into bot applications.