English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

When recognizing emotions from speech, we encounter two common problems: howto optimally capture emotion-relevant information from the speech signal andhow to best quantify or categorize the noisy subjective emotion labels.Self-supervised pre-trained representations can robustly capture informationfrom speech enabling state-of-the-art results in many downstream tasksincluding emotion recognition. However, better ways of aggregating theinformation across time need to be considered as the relevant emotioninformation is likely to appear piecewise and not uniformly across the signal.For the labels, we need to take into account that there is a substantial degreeof noise that comes from the subjective human annotations. In this paper, wepropose a novel approach to attentive pooling based on correlations between therepresentations' coefficients combined with label smoothing, a method aiming toreduce the confidence of the classifier on the training labels. We evaluate ourproposed approach on the benchmark dataset IEMOCAP, and demonstrate highperformance surpassing that in the literature. The code to reproduce theresults is available at github.com/skakouros/s3prl_attentive_correlation.

Speech-based emotion recognition with self-supervised models using attentive channel-wise correlations and label smoothing