Technical Name Personalized emotion sensing for spoken dialog interface
Operator National Tsing Hua University
Booth AI&IOT applications
Technical Description The technology is an integrated solution incorporating the retrievable personality for multimodal emotion sensing for spoken services. This framework provides a real-world flexibility that enables the estimation of the target speaker emotion states without manual personalization.
Scientific Breakthrough When considering individuality in advancing emotion recognition, better modeling individual differences of human emotion expression is key. Our algorithm computes an unsupervised individual spaceachieves the best accuracy in the emotion benchmark. The work is published in the flagship conferences on affective computing (ACII2019)the worlds largest speech technology (INTERSPEECH2019).
Industrial Applicability User interface using speech technology will benefit from such a technology (SIRI, Alexa). Other applications can be found in customer care center analytics, remote human resource recruitment,even affect disorder health application.
Contact 林子晴