Abstract
Most of the studies on speech emotion recognition have used single-language corpora, but little research has been done in cross-language valence speech emotion recognition. Research has shown that the models developed for single-language speech recognition systems perform poorly when used in different environments. Cross-language speech recognition is a craving alternative, but it is highly challenging because the corpora used will have been recorded in different environments and under varying conditions. The differences in the quality of recording devices, elicitation techniques, languages, and accents of speakers make the recognition task even more arduous. In this paper, we propose a stacked ensemble learning algorithm to recognize valence emotion in a cross-language speech environment. The proposed ensemble algorithm was developed from random decision forest, AdaBoost, logistic regression, and gradient boosting machine and is therefore called RALOG. In addition, we propose feature scaling using random forest recursive feature elimination and a feature selection algorithm to boost the performance of RALOG. The algorithm has been evaluated against four widely used ensemble algorithms to appraise its performance. The amalgam of five benchmarked corpora has resulted in a cross-language corpus to validate the performance of RALOG trained with the selected acoustic features. The comparative analysis results have shown that RALOG gave better performance than the other ensemble learning algorithms investigated in this study.
Highlights
The advent of machine learning methods has brought about groundbreaking research and ubiquitous applications of different algorithms to solve challenging problems facing humans.Nowadays, we live in a world where the idea of detecting skin melanoma [1] and plant diseases [2] is made possible because of the innovation of machine learning
The authors of the present study have investigated the problem of cross-language valence speech emotion recognition for different languages documented in the five corporal of EMO-DB, Surrey audio-visual expressed emotion (SAVEE), RAVDESS, EMOVO, and crowd-sourced emotional multimodal acted dataset (CREMA-D)
Negative and positive valence emotion files were extracted from the speech corpora of EMO-DB, SAVEE, RAVDESS, EMOVO, and CREMA-D to create the cross-language corpus according of EMO-DB, SAVEE, RAVDESS, EMOVO, and CREMA-D to create the cross-language corpus to the mapping in [25]
Summary
The advent of machine learning methods has brought about groundbreaking research and ubiquitous applications of different algorithms to solve challenging problems facing humans.Nowadays, we live in a world where the idea of detecting skin melanoma [1] and plant diseases [2] is made possible because of the innovation of machine learning. The advent of machine learning methods has brought about groundbreaking research and ubiquitous applications of different algorithms to solve challenging problems facing humans. The field of communication network has not been spared in the pervasive adoption of intelligent systems for supporting human decision-making process. The necessity to develop intelligent robotic systems has triggered innovative research in many fields such as the healthcare, industry automation [3], human–computer interaction [4], and many other practical application areas. The development of a model that can recognize customer satisfaction requires a blend of different languages and accents. Developing systems that are emotionally intelligent requires an appropriate model to represent emotions [6]. The categorical approach and dimensional approach are two possible ways of modeling human emotions [7]. The dimensional approach presents speech emotions in different dimensions such as the three dimensions of arousal, dominance, and valence [8]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
