A new hybrid PSO assisted biogeography-based optimization for emotion and stress recognition from speech signal

作者：

Highlights：

• OSBSBCFs were used for multiclass emotion/stress recognition from speech signal.

• A New Hybrid PSO Assisted BBO (PSOBBO) was proposed for feature selection.

• Simulations were conducted for three speech emotion and also validated using eight benchmark datasets.

• The best prediction performances were achieved for the simulations conducted.

摘要

•OSBSBCFs were used for multiclass emotion/stress recognition from speech signal.•A New Hybrid PSO Assisted BBO (PSOBBO) was proposed for feature selection.•Simulations were conducted for three speech emotion and also validated using eight benchmark datasets.•The best prediction performances were achieved for the simulations conducted.

论文关键词：Speech signals,Emotions,Feature extraction,Feature selection and emotion recognition

论文评审过程：Received 11 May 2016, Revised 25 August 2016, Accepted 16 October 2016, Available online 17 October 2016, Version of Record 26 October 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2016.10.035