Cross-corpus speech emotion recognition, which learns an accurate classifier for new test data using old and labeled training data, has shown promising value in speech emotion recognition research. Most previous works have explored two learning strategies independently for cross-corpus speech emotion recognition: feature matching and sample reweighting. In this paper, we show that both strategies are important and inevitable when the distribution difference is substantially large for training and test data. We therefore put forward a novel multiple kernel learning of joint sample and feature matching (JSFM-MKL) to model them in a unified optimization problem. Experimental results demonstrate that the proposed JSFM-MKL outperforms the competitive algorithms for cross-corpus speech emotion recognition.
from #AlexandrosSfakianakis via Alexandros G.Sfakianakis on Inoreader http://ift.tt/2xHHguP
via IFTTT
Εγγραφή σε:
Σχόλια ανάρτησης (Atom)
Δημοφιλείς αναρτήσεις
-
Abstract Objective The objective of this study were to find the annual case trend of inflammatory jawbone diseases and to investigate th...
-
The new app from IBM will let crowds automatically coordinates mass displays at big sporting events or protests using their smartphone camer...
-
The investigational third-generation nonsteroidal oral selective estrogen receptor degrader RAD1901 was associated with a 23% objective resp...
-
by Patrick P. Lowe, Benedek Gyongyosi, Abhishek Satishchandran, Arvin Iracheta-Vellve, Aditya Ambade, Yeonhee Cho, Karen Kodys, Donna Catala...
-
A magnetic Fe3O4/GO composite with potential for rapid solid-liquid separation through a magnetic field was synthesized using GO (graphene o...
-
Very high cycle fatigue behavior of nugget zone in AA 7075 friction stir welded joint was experimentally investigated using ultrasonic fatig...
-
People with attention deficit hyperactivity disorder (ADHD) had a lower risk of motor vehicle crashes when they had received their medicatio...
-
Abstract Glutamine (GLUT) is a nonessential amino acid that can become conditionally essential under stress conditions, being able to act ...
-
Abstract Purpose To determine the histological distribution and trends in incidence of sinonasal malignancies in Jamaica. Methods Ca...
-
from #AlexandrosSfakianakis via Alexandros G.Sfakianakis on Inoreader http://ift.tt/2qS0o5m via IFTTT
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου