Learn More
The INTERSPEECH 2014 Computational Paralinguistics Challenge provides for the first time a unified test-bed for the automatic recognition of speakers' cognitive and physical load in speech. In this paper, we describe these two Sub-Challenges, their conditions, baseline results and experimental procedures, as well as the COMPARE baseline features generated(More)
Image mosaic or image montage is an image made up of many other images. An image mosaic has its own visual content as a whole while each of its building images also has a meaningful content. This thesis address the design and implementation of an image mosaic generating system, which uses computers to build image mosaics automatically. There are many(More)
In this contribution, we propose a novel method for Active Learning (AL) - <i>Dynamic Active Learning (DAL)</i> - which targets the reduction of the costly human labelling work necessary for modelling subjective tasks such as emotion recognition in spoken interactions. The method implements an adaptive query strategy that minimises the amount of human(More)
Emotion representations are psychological constructs for modelling, analysing, and recognising emotion, being one essential element of affect. Due to its complexity, the boundaries between different emotion concepts are often fuzzy, which is also reflected in the diversification of emotion databases, and their inconsistent target labels. When facing data(More)
In this work, we present an in-depth analysis of the interdependency between the non-native prosody and the native language (L1) of English L2 speakers, as separately investigated in the Degree of Nativeness Task and the Native Language Task of the INTERSPEECH 2015 and 2016 Computational Paralinguistics ChallengE (ComParE). To this end, we propose a(More)