Excitation Features of Speech for Speaker-Specific Emotion Detection
In this article, we study emotion detection from speech in a speaker-specific scenario. By parameterizing the excitation component of voiced speech, the study explores deviations between emotional speech (e.g., speech produced in anger, happiness, sadness, etc.) and neutral speech (i.e., non-emotion...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9046041/ |
id |
doaj-bab65234fb174c14bb51b49f788381a6 |
---|---|
record_format |
Article |
spelling |
doaj-bab65234fb174c14bb51b49f788381a62021-03-30T01:30:48ZengIEEEIEEE Access2169-35362020-01-018603826039110.1109/ACCESS.2020.29829549046041Excitation Features of Speech for Speaker-Specific Emotion DetectionSudarsana Reddy Kadiri0https://orcid.org/0000-0001-5806-3053Paavo Alku1Department of Signal Processing and Acoustics, Aalto University, Espoo, FinlandDepartment of Signal Processing and Acoustics, Aalto University, Espoo, FinlandIn this article, we study emotion detection from speech in a speaker-specific scenario. By parameterizing the excitation component of voiced speech, the study explores deviations between emotional speech (e.g., speech produced in anger, happiness, sadness, etc.) and neutral speech (i.e., non-emotional) to develop an automatic emotion detection system. The excitation features used in this study are the instantaneous fundamental frequency, the strength of excitation and the energy of excitation. The Kullback-Leibler (KL) distance is computed to measure the similarity between feature distributions of emotional and neutral speech. Based on the KL distance value between a test utterance and an utterance produced in a neutral state by the same speaker, a detection decision is made by the system. In the training of the proposed system, only three neutral utterances produced by the speaker were used, unlike in most existing emotion recognition and detection systems that call for large amounts of training data (both emotional and neutral) by several speakers. In addition, the proposed system is independent of language or lexical content. The system is evaluated using two databases of emotional speech. The performance of the proposed detection method is shown to be better than that of reference methods.https://ieeexplore.ieee.org/document/9046041/Speech analysisparalinguisticsemotion detectionexcitation sourcezero frequency filtering (ZFF)linear prediction (LP) analysis |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Sudarsana Reddy Kadiri Paavo Alku |
spellingShingle |
Sudarsana Reddy Kadiri Paavo Alku Excitation Features of Speech for Speaker-Specific Emotion Detection IEEE Access Speech analysis paralinguistics emotion detection excitation source zero frequency filtering (ZFF) linear prediction (LP) analysis |
author_facet |
Sudarsana Reddy Kadiri Paavo Alku |
author_sort |
Sudarsana Reddy Kadiri |
title |
Excitation Features of Speech for Speaker-Specific Emotion Detection |
title_short |
Excitation Features of Speech for Speaker-Specific Emotion Detection |
title_full |
Excitation Features of Speech for Speaker-Specific Emotion Detection |
title_fullStr |
Excitation Features of Speech for Speaker-Specific Emotion Detection |
title_full_unstemmed |
Excitation Features of Speech for Speaker-Specific Emotion Detection |
title_sort |
excitation features of speech for speaker-specific emotion detection |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2020-01-01 |
description |
In this article, we study emotion detection from speech in a speaker-specific scenario. By parameterizing the excitation component of voiced speech, the study explores deviations between emotional speech (e.g., speech produced in anger, happiness, sadness, etc.) and neutral speech (i.e., non-emotional) to develop an automatic emotion detection system. The excitation features used in this study are the instantaneous fundamental frequency, the strength of excitation and the energy of excitation. The Kullback-Leibler (KL) distance is computed to measure the similarity between feature distributions of emotional and neutral speech. Based on the KL distance value between a test utterance and an utterance produced in a neutral state by the same speaker, a detection decision is made by the system. In the training of the proposed system, only three neutral utterances produced by the speaker were used, unlike in most existing emotion recognition and detection systems that call for large amounts of training data (both emotional and neutral) by several speakers. In addition, the proposed system is independent of language or lexical content. The system is evaluated using two databases of emotional speech. The performance of the proposed detection method is shown to be better than that of reference methods. |
topic |
Speech analysis paralinguistics emotion detection excitation source zero frequency filtering (ZFF) linear prediction (LP) analysis |
url |
https://ieeexplore.ieee.org/document/9046041/ |
work_keys_str_mv |
AT sudarsanareddykadiri excitationfeaturesofspeechforspeakerspecificemotiondetection AT paavoalku excitationfeaturesofspeechforspeakerspecificemotiondetection |
_version_ |
1724186944387153920 |