Microsoft launches 4 TTS models including zh-CN-YunjieNeural, specially designed for AI chat

IT House News on September 22, Microsoft stated that as generative AI becomes popular all over the world, during the chat process between users and AI, it also puts forward higher requirements for the naturalness and expressiveness of text-to-speech.

Microsoft today announced the launch of four TTS voices: en-US-AndrewNeural, en-US-BrianNeural, en-US-EmmaNerual and zh-CN-YunjieNeural. They are currently available in public preview in three markets: the Eastern United States, Southeast Asia and Western Europe. .

Microsoft officially stated that the four TTS voices can provide natural voice interaction, making chat sounds closer to reality, more natural, and more attractive.

In the press release, Microsoft also showed the comparative effects of four TTS voices and demonstrated different tone styles. IT House has attached a link here. Interested users can click on the link to experience it in depth

Microsoft said that developers can leverage the Azure Speech SDK and REST API to integrate speech into existing applications through Azure OpenAI

The above is the detailed content of Microsoft launches 4 TTS models including zh-CN-YunjieNeural, specially designed for AI chat. For more information, please follow other related articles on the PHP Chinese website!