4 datasets found

The MSP-IMPROV dataset contains 6 sessions of dyadic interactions between pairs of male-female actors. 15 target sentences are used to collect the recordings. For each target...

We propose a deep graph approach to address the task of speech emotion recognition. A compact, efficient and scalable way to repre- sent data is in the form of graphs.

The Connectionist Temporal Classification dataset is a speech emotion recognition dataset, consisting of audio recordings of speech.

Speaker verification systems experience significant performance degradation when tasked with short-duration trial recordings. To address this challenge, a multi-scale feature...