Hong Kong Applied Science and Technology Research Institute Company, Limited

Ho Pong Sze

Kar-Wing Edward Lor

Ironside Hoi Yeung Lam

Chun Chung Ho

A method for training a speech-emotion recognition classifier under a continuously updatable and re-trainable ASER machine learning model. The quantified training data is generated by first processing the utterances of a human speech source and the associated texts in an emotion evaluation and rating process with normalization; then, extracting the features of the utterance; quantifying the feature attributes of the extracted features by labelling, tagging, and weighting the feature attributes, with their values assigned under measurable scales. The quantified training data comprises the normalized results of the emotion evaluation and rating process, the extracted features, the quantified emotional feature attributes, and the hash values of the quantified emotional feature attributes.

Apparatus and method for speech-emotion recognition with quantified emotional states

The present invention includes apparatuses, systems, and methods for peer group formation for mobile devices by proximity sensing. The described embodiments further include methods, systems, and apparatuses configured to easily establish an ad-hoc peer network in which each of the devices in the network automatically discovers and is connected to other devices in the network upon establishing a connection with at least one member of the ad-hoc network. The connection is established using a tap of NFC devices in the two wireless devices. Further, the wireless devices in the ad-hoc network may be divided into logical peer groups.

Growing community of inventors

Hong Kong, China

Ho Pong Sze