Speech Quality Enhancement in Digital Forensic Voice Analysis

Ekpenyong, Moses E.; Obot, Okure U.

doi:10.1007/978-3-319-05885-6_18

Cited by 1 publication

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of course, the proposed architecture can be integrated into call center departments to analyze agent conversations and determine their as well as their customers' emotions, providing advice on how to improve the customer satisfaction and quality of the service [22,23]. Analyzing the speech from phone conversations or suspect interrogation and studying the emotional patterns can also help crime investigation departments to find the perpetrator [28,51]. Nonetheless, analyzing 911 calls can determine the genuineness of the call and flag possible false calls as well as sort the calls based on the emergency level predicted based on callers' speech [28].…”

Section: Discussionmentioning

confidence: 99%

“…The proposed SER system can be successfully used in HCI based on speech interaction, as well as integrated in computer-mediated education systems [47], for airplane or car accidents prevention [30,31], diagnosing mental disorders [24][25][26] and physical diseases [27], in call centers to improve the quality of customer service [22,23], and crime investigation departments by analyzing emotional patterns during suspect interrogations to find the perpetrator [28,51].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Feedforward Neural Network-Based Architecture for Predicting Emotions from Speech

Gavrilescu

Vizireanu

2019

Data

View full text Add to dashboard Cite

We propose a novel feedforward neural network (FFNN)-based speech emotion recognition system built on three layers: A base layer where a set of speech features are evaluated and classified; a middle layer where a speech matrix is built based on the classification scores computed in the base layer; a top layer where an FFNN- and a rule-based classifier are used to analyze the speech matrix and output the predicted emotion. The system offers 80.75% accuracy for predicting the six basic emotions and surpasses other state-of-the-art methods when tested on emotion-stimulated utterances. The method is robust and the fastest in the literature, computing a stable prediction in less than 78 s and proving attractive for replacing questionnaire-based methods and for real-time use. A set of correlations between several speech features (intensity contour, speech rate, pause rate, and short-time energy) and the evaluated emotions is determined, which enhances previous similar studies that have not analyzed these speech features. Using these correlations to improve the system leads to a 6% increase in accuracy. The proposed system can be used to improve human–computer interfaces, in computer-mediated education systems, for accident prevention, and for predicting mental disorders and physical diseases.

show abstract