Analysis of Deep Learning-Based Decision-Making in an Emotional Spontaneous Speech Task
In this work, we present an approach to understand the computational methods and decision-making involved in the identification of emotions in spontaneous speech. The selected task consists of Spanish TV debates, which entail a high level of complexity as well as additional subjectivity in the human...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-01-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/13/2/980 |
_version_ | 1797446555814330368 |
---|---|
author | Mikel de Velasco Raquel Justo Asier López Zorrilla María Inés Torres |
author_facet | Mikel de Velasco Raquel Justo Asier López Zorrilla María Inés Torres |
author_sort | Mikel de Velasco |
collection | DOAJ |
description | In this work, we present an approach to understand the computational methods and decision-making involved in the identification of emotions in spontaneous speech. The selected task consists of Spanish TV debates, which entail a high level of complexity as well as additional subjectivity in the human perception-based annotation procedure. A simple convolutional neural model is proposed, and its behaviour is analysed to explain its decision-making. The proposed model slightly outperforms commonly used CNN architectures such as VGG16, while being much lighter. Internal layer-by-layer transformations of the input spectrogram are visualised and analysed. Finally, a class model visualisation is proposed as a simple interpretation approach whose usefulness is assessed in the work. |
first_indexed | 2024-03-09T13:43:11Z |
format | Article |
id | doaj.art-52f9b8c87551460eb2113493259e7f8b |
institution | Directory Open Access Journal |
issn | 2076-3417 |
language | English |
last_indexed | 2024-03-09T13:43:11Z |
publishDate | 2023-01-01 |
publisher | MDPI AG |
record_format | Article |
series | Applied Sciences |
spelling | doaj.art-52f9b8c87551460eb2113493259e7f8b2023-11-30T21:04:27ZengMDPI AGApplied Sciences2076-34172023-01-0113298010.3390/app13020980Analysis of Deep Learning-Based Decision-Making in an Emotional Spontaneous Speech TaskMikel de Velasco0Raquel Justo1Asier López Zorrilla2María Inés Torres3Department of Electricity and Electronics, Faculty of Science and Technology, University of the Basque Country UPV/EHU, 48940 Leioa, SpainDepartment of Electricity and Electronics, Faculty of Science and Technology, University of the Basque Country UPV/EHU, 48940 Leioa, SpainDepartment of Electricity and Electronics, Faculty of Science and Technology, University of the Basque Country UPV/EHU, 48940 Leioa, SpainDepartment of Electricity and Electronics, Faculty of Science and Technology, University of the Basque Country UPV/EHU, 48940 Leioa, SpainIn this work, we present an approach to understand the computational methods and decision-making involved in the identification of emotions in spontaneous speech. The selected task consists of Spanish TV debates, which entail a high level of complexity as well as additional subjectivity in the human perception-based annotation procedure. A simple convolutional neural model is proposed, and its behaviour is analysed to explain its decision-making. The proposed model slightly outperforms commonly used CNN architectures such as VGG16, while being much lighter. Internal layer-by-layer transformations of the input spectrogram are visualised and analysed. Finally, a class model visualisation is proposed as a simple interpretation approach whose usefulness is assessed in the work.https://www.mdpi.com/2076-3417/13/2/980emotion detectionspeech processingexplainable artificial intelligencemachine learning |
spellingShingle | Mikel de Velasco Raquel Justo Asier López Zorrilla María Inés Torres Analysis of Deep Learning-Based Decision-Making in an Emotional Spontaneous Speech Task Applied Sciences emotion detection speech processing explainable artificial intelligence machine learning |
title | Analysis of Deep Learning-Based Decision-Making in an Emotional Spontaneous Speech Task |
title_full | Analysis of Deep Learning-Based Decision-Making in an Emotional Spontaneous Speech Task |
title_fullStr | Analysis of Deep Learning-Based Decision-Making in an Emotional Spontaneous Speech Task |
title_full_unstemmed | Analysis of Deep Learning-Based Decision-Making in an Emotional Spontaneous Speech Task |
title_short | Analysis of Deep Learning-Based Decision-Making in an Emotional Spontaneous Speech Task |
title_sort | analysis of deep learning based decision making in an emotional spontaneous speech task |
topic | emotion detection speech processing explainable artificial intelligence machine learning |
url | https://www.mdpi.com/2076-3417/13/2/980 |
work_keys_str_mv | AT mikeldevelasco analysisofdeeplearningbaseddecisionmakinginanemotionalspontaneousspeechtask AT raqueljusto analysisofdeeplearningbaseddecisionmakinginanemotionalspontaneousspeechtask AT asierlopezzorrilla analysisofdeeplearningbaseddecisionmakinginanemotionalspontaneousspeechtask AT mariainestorres analysisofdeeplearningbaseddecisionmakinginanemotionalspontaneousspeechtask |