Does the chat robot have an expression?

Now, a new patent granted to Microsoft by the United States Patent and Trademark Office (USPTO) shows that the company is developing a dialogue agent that can reflect the dialogue style and/or facial expressions of users.

It is reported that the title of this application is language style matching agent, which was obtained on September 3, 2020.

USPTO approval. Daniel J. Macduff, Kyle R. Tam, Mary P. Chevinski, Dipari Aneha and Rennes.

Hogan is regarded as the inventor.

This new patented technology aims to bring an end-to-end voice-based dialogue agent, which can conduct more natural and multi-round dialogues and keep consistent with the user's dialogue style and facial expressions.

By recognizing the prosodic features of users' voices and synthesizing voices for virtual subjects with the same or similar features, the language style matching of dialogue subjects can be realized. The user's facial expression can be recognized and imitated by the face of a specific conversation agent. The predetermined response can be generated by a combination of script-based virtual machine responses.

This also means that a dialogue agent consistent with the user's dialogue style and facial expression may be considered more trustworthy, easier to understand and create a more natural human-computer interaction. By using machine learning technology to respond to users' words in a natural and understandable way, human-computer interaction will also be improved as a whole.