Machines that read emotion

A new system created by Spanish researchers has enabled machines to adapt their response to suit users’ emotions. Focusing on the negative emotions usually experienced when interacting with an automatic system – namely anger, boredom and doubt - scientists have designed the model to adapt to the tone of the conversation and make dialogue faster and more effective.

Around sixty different acoustic parameters were gathered including tone of voice, speed of speech, duration of pauses and the energy of the voice signal, in order for the machine to determine both the emotion and the intention of the user. The machine then progresses the dialogue in the most appropriate manner. For example, if the machine detects ‘doubt’ it may be that further detailed instructions would be required, a response that would provoke a negative reaction if used when the user is emitting indications of ‘anger’.

Such a system is perceived as being more useful by users and on average makes conversations shorter. The study was carried out by Professor David Grill Barres, of the Applied Artificial Intelligence Group of UC3M’s Computer Science Department, together with Professors Zoraida Callejas Carrión and Ramón López-Cózar Delgado, of the Spoken and Multimodal Dialogue Group of the Computer Languages and Systems Department of the UGR.

Click here to see the full press release.

Published: Monday, 21st November 2011 by Ellen Haggan