From video game characters, chatbots, content readers to customer service, user assistance,..., the more voice emotions and styles support, the better end-user engaging experience delivering. Understanding those needs, we decided to add many emotions and styles to a batch of 10 neural voices in US English, French, Japanese and Mandarin Chinese.
Each voice will support a range of emotional tones, from cheerful, angry, sad, excited to hopeful, friendly, unfriendly, and terrified. Some voices also support shouting, whispering, customer service and newscast options.
The new feature is currently in beta status and available for subscribed users only. From their feedback, we could add more voices to the support list, add or remove emotions on current voices,..., all depending on customer satisfaction.
By supporting more voice options with emotions and styles, our text to speech online service continues to address the unmet needs of the customers to build more delightful speech experience. Hopefully we can open the feature to all our customers including free users in future.