Zhipu AI has announced that its end-to-end emotional speech technology is now officially available on the Zhipu Chat platform, open to all users. This technology breaks through the limitations of traditional text-to-speech (TTS) technology, enabling a deep understanding of contextual nuances and generating emotionally rich natural conversations. This marks a significant evolution in Zhipu AI's voice synthesis technology from simple text reading to an AI capable of expressing genuine emotions.
The emotional speech technology of Zhipu Chat boasts a variety of advanced features. It not only understands and expresses various emotions in speech, such as joy, anger, sorrow, and happiness, but also supports multiple languages and dialects, including Cantonese, Northeastern Mandarin, and foreign languages like English and Japanese. Additionally, users can interrupt the voice output at any time and adjust parameters such as volume and speech speed flexibly. Notably, these features can be combined within a single sentence, making the voice output more vivid, rich, and warm, akin to a real person.
Currently, the emotional speech feature is fully launched on the Zhipu Chat app, allowing users to immediately experience this functionality without waiting. To use this feature, users simply need to update the Zhipu Chat app to the latest version and click the dialogue button in the bottom right corner of the chat box to interact with Xiao Zhi and feel its unique charm.