Whispered Speech Emotion Recognition with Gender Detection using BiLSTM and DCNN

Title

Whispered Speech Emotion Recognition with Gender Detection using BiLSTM and DCNN

Creator

Mohanty A.; Cherukuri R.C.

Description

Emotions are human mental states at a particular instance in time concerning ones circumstances, mood, and relationships with others. Identifying emotions from the whispered speech is complicated as the conversation might be confidential. The representation of the speech relies on the magnitude of its information. Whispered speech is intelligible, a low-intensity signal, and varies from normal speech. Emotion identification is quite tricky from whispered speech. Both prosodic and spectral speech features help to identify emotions. The emotion identification in a whispered speech happens using prosodic speech features such as zero-crossing rate (ZCR), pitch, and spectral features that include spectral centroid, chroma STFT, Mel scale spectrogram, Mel-frequency cepstral coefficient (MFCC), Shifted Delta Cepstrum (SDC), and Spectral Flux. There are two parts to the proposed implementation. Bidirectional Long Short-Term Memory (BiLSTM) helps to identify the gender from the speech sample in the first step with SDC and pitch. The Deep Convolutional Neural Network (DCNN) model helps to identify the emotions in the second step. This implementation is evaluated using the wTIMIT data corpus and gives 98.54% accuracy. Emotions have a dynamic effect on genders, so this implementation performs better than traditional approaches. This approach helps to design online learning management systems, different applications for mobile devices, checking cyber-criminal activities, emotion detection for older people, automatic speaker identification and authentication, forensics, and surveillance. (2023), (Iranian Academic Center for Education). All Rights Reserved.

Source

Journal of Information Systems and Telecommunication, Vol-12, No. 2, pp. 152-161.

Date

2024-01-01

Publisher

Iranian Academic Center for Education, Culture and Research

Subject

BiLSTM; Data Corpus; DCN; Emotion Recognition; Speech Features; Whispered Speech

Coverage

Mohanty A., Department of Computer science and Engineering, Christ (Deemed to be University), Bangalore, India; Cherukuri R.C., Department of Computer science and Engineering, Christ (Deemed to be University), Bangalore, India

Rights

Restricted Access

Relation

ISSN: 23221437

Format

Online

Language

English

Type

Article

Identifier

https://doi.org/10.61186/jist.43703.12.46.152

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85201782931&doi=10.61186%2fjist.43703.12.46.152&partnerID=40&md5=d057298b375b6bde82e9fad2487d0505