In September, Microsoft revealed it has created several realistic-sounding AI text-to-speech voices for customers of its Azure AI services, three in US English, and one in Chinese. Today, the company announced seven new AI TTS voices for those users, along with updates to previously released voices, via a public preview.
In a blog post, Microsoft stated:
We are introducing these new voices specifically designed for conversational scenarios. Whether you are creating a speech-based chatbot, a voice assistant, or a conversational agent, these new voices will ensure your interactions are more realistic, lifelike, and engaging. The new realistic conversational voices are perfect matches for any application necessitating lifelike speech interactions, including chatbots, voice assistants, gaming, e-learning, entertainment, and more.
Microsoft previously mentioned how it developed these new more realistic-sounding AI voices:
We began by crafting the persona of each voice as if it were a real person who is friendly and optimistic about life, always eager to assist others and share intriguing or practical knowledge. The speaking style of the voice resembles a conversation with an acquaintance over a cup of tea, maintaining a natural and unexaggerated tone.
The new voices are as follows
- German (Germany) - Female - Seraphina
- Spanish (Spain) - Female- Ximena
- French (Canada) - Male - Thierry
- French (France) - Female - Vivienne
- Italian (Italy) - Male - Giuseppe
- Korean (Korea) - Male - Hyuns
- Portuguese (Brazil) - Female - Thalita
In addition, the previously released Xiaoxiao female Mandarin Chinese (China) voice has been updated with a "more natural speaking style". Also, three other voices have been updated as well for "more expressive prosody"
- Spanish (Spain) - Male - Alvaro
- English (United Kingdom) - Male - Ryan
- Korean (Korea) - Male - Injoon
You can check out samples for all the new and updated Azure text-to-speech voices on the blog post.
Microsoft already offers over 400 neural voices that cover over 140 languages and locales via Azure. The company also offers business customers a way to make their own unique digital spokesperson with its Custom Neural Voice feature.