Novel cascaded Gaussian mixture model-deep neural network classifier for speaker identification in emotional talking environments

Abstract

This research is an effort to present an effective approach to enhance text-independent speaker identification performance in emotional talking environments based on novel classifier called cascaded Gaussian Mixture Model-Deep Neural Network (GMM-DNN). Our current work focuses on proposing, implementing and evaluating a new approach for speaker identification in emotional talking environments based on cascaded Gaussian Mixture Model-Deep Neural Network as a classifier. The results point out that the cascaded GMM-DNN classifier improves speaker identification performance at various emotions using two distinct speech databases:Emirati speech database (Arabic United Arab Emirates dataset) and "Speech Under Simulated and Actual Stress (SUSAS)" English dataset. The proposed classifier outperforms classical classifiers such as Multilayer Perceptron (MLP) and Support Vector Machine (SVM) in each dataset. Speaker identification performance that has been attained based on the cascaded GMM-DNN is similar to that acquired from subjective assessment by human listeners.2 Keywords: deep neural network; emotional talking environments; Gaussian mixture model; speaker identification. IntroductionSpeaker recognition and its sub divisional entities: "speaker identification and speaker verification" need to be redefined on the basis of talking environments as the neutral talking environment and the emotional talking environments. Speaker recognition performance faces drastic challenges, especially when the speaker identity going through the human-computer interface in emotional talking environments. Speaker recognition applications in security systems are widening their base into banking sector, customer care sector, criminal investigation and can be used as security control measure to remotely access a server or for access to confidential library files on a server. The process of "automatic speaker identification and verification" in stressful and emotional talking environments is a challenging area of research [1]. "Speaker identification" is comprised of two schemes in terms of sets: "closed set" and "open set" speaker identification. When the unknown speaker is presumed to be one among the database of known speakers, it becomes the scheme of a "closed set", while in the scheme of an "open set", the unfamiliar speaker might not necessarily be from the database of familiar speakers. Operational procedure divides speaker identification into "text-dependent", where the same text is uttered by the speaker in the training and testing phases and "text-independent", where different texts are uttered by the speaker during the training and testing phases [2]. 3 A perfect communication from a speaker depends not only on linguistic statements but also on the emotional aspects of the speaker. Identifying the emotional aspects of the speaker by the machine is still a challenge of the human-machine interface. Speech is always a perfect mix of linguistic notes linked with emotion along with its paralinguistic features. Emotion recognition from ...

Speaker Identification in Different Emotional States in Arabic and English

“…There is indeed a very small body of work on SI for Arabic especially in an emotional state as shown in Table 1. Table 8 presents a comparison between our CRNN model and Shahin's models [28], [30], [44], [45], and [60] to identify a speaker in a specific emotion. Table 9 presents another comparison between the proposed system and other previous studies in an emotional environment.…”

“…Table 1 presents a summary of Arabic speaker recognition from 2010 to 2019 in different environments of neutral and emotional speech, as well as shouting. Many remarkable studies were reported by Shahin et al [6], [28]- [30] with the goal of developing SV systems using different classifiers and targeting Emirati-accented speakers. Shahin et al [30] proposed a novel classifier called the cascaded GMM deep neural network (GMM-DNN) to enhance text-independent SI performance using two corpora: the Emirati speech database and ''speech under simulated and actual stress'' English database.…”

“…Many remarkable studies were reported by Shahin et al [6], [28]- [30] with the goal of developing SV systems using different classifiers and targeting Emirati-accented speakers. Shahin et al [30] proposed a novel classifier called the cascaded GMM deep neural network (GMM-DNN) to enhance text-independent SI performance using two corpora: the Emirati speech database and ''speech under simulated and actual stress'' English database. Additionally, Shahin et al [29] identified Emirati-accented speakers in neutral talking and shouting environments.…”

Emirati-Accented Speaker Identification in Stressful Talking Conditions

“…In a recent study by Shahin et.al [7], they proposed an efficient methodology to improve "text-independent speaker identification accuracy in emotional talking environments based on a novel classifier called cascaded Gaussian Mixture Model-Deep Neural Network (GMM-DNN). Specifically, their work focused on proposing, implementing, and assessing a new framework for speaker identification in emotional talking environments based on cascaded Gaussian Mixture Model-Deep Neural Network as a classifier.…”

Speaker Verification in Emotional Talking Environments based on Third-Order Circular Suprasegmental Hidden Markov Model

“…In emotional/stressful talking environments, Shahin et al [5] enhanced "text-independent speaker identification accuracy in emotional talking environments based on novel classifier named cascaded Gaussian Mixture Model-Deep Neural Network (GMM-DNN). Their work focused on proposing, applying, and testing a new framework for speaker identification in such talking environments based on sequential Gaussian Mixture Model followed by Deep Neural Network as a classifier.…”