Prediction of Speech Quality Based on Resilient Backpropagation Artificial Neural Network

Orčík, Lukáš; Vozňák, Miroslav; Rozhon, Jan; Rezac, Filip; Slachta, Jiri; Toral-Cruz, Homero; Lin, Jerry Chun‐Wei

doi:10.1007/s11277-016-3746-2

Cited by 8 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is worth mentioning that when it comes to designing and evaluating digital broadcasting systems, the process of frequency allotment, resource allocation, and ensemble configuration is still a widely discussed topic [31,32]. Advances in audio quality assessment are discussed in [33][34][35][36][37][38][39][40][41]. Until now, only Norway performed the so-called switchover, leaving analog FM radio behind.…”

Section: Related Workmentioning

confidence: 99%

Digital Transformation of Terrestrial Radio: An Analysis of Simulcasted Broadcasts in FM and DAB+ for a Smart and Successful Switchover

Falkowski-Gilski

2021

Applied Sciences

View full text Add to dashboard Cite

The process of digitizing radio is far from over. It is an important interdisciplinary aspect, involving Big Data and AI (Artificial Intelligence) when it comes to classifying and handling content, and an organizational challenge in the Industry 4.0 concept. There exist several methods for delivering audio signals, including terrestrial broadcasting and internet streaming. Among them, the DAB+ (Digital Audio Broadcasting plus) system is one of the leading standards of terrestrial digital radio transmission. Compared with analog FM (frequency modulation) radio, it is more bandwidth efficient and offers greater possibilities when it comes to delivering content and forming an ensemble and multiplex. Currently, many countries worldwide, particularly European States, are still making adjustments in order to perform an efficient switchover from analog FM to digital DAB+ radio. This paper presents the current situation of the digital radio market as well as the results of a subjective quality evaluation study and questionnaire concerning broadcasting in both digital and analog techniques. It involves radio programs, transmitting both speech and music signals, simulcasted in DAB+ and FM standards. It also presents the development of the national multiplex. The results of this study may help both researchers and scientists as well as policy makers and professionals active in the field of broadcasting and electronic media and not to mention the consumption of multimedia content.

show abstract

Section: Related Workmentioning

confidence: 99%

Digital Transformation of Terrestrial Radio: An Analysis of Simulcasted Broadcasts in FM and DAB+ for a Smart and Successful Switchover

Falkowski-Gilski

2021

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Currently, the most commonly used objective audio quality predictors, PEAQ (Perceptual Evaluation of Audio Quality) [5] (or PESQ Perceptual Evaluation of Speech Quality), POLQA (Perceptual Objective Listening Quality Analysis) [6], PEMO-Q (Perception Model for Quality Assessment) [7], STOI (Short-Time Objective Intelligibility) [8], VISQOL (Virtual Speech Quality Objective Listener) [9], SNR (Signal-to-Noise Ratio) [10] require the reference signals and/or specific types of noise that may degrade audio quality [11]. Existing non-reference solutions [12] are mainly focused on quality measurements of speech (ANIQUE (Auditory Non-Intrusive Quality Estimation) [13], HASQI (Hearing Aid Speech Quality Index) [14], POSQE (Perceptual Output-based Speech Quality Evaluation) [15], SRMR (Standardized Root Mean Square) [16] and others [17,18]), synthetic audio signals [19], image [20] or video [19,21,22]. To the best of the authors' knowledge, there is no recommended non-reference objective method for a real music quality assessment.…”

Section: The Artefacts Detection In Audio Signals -State Of the Artmentioning

confidence: 99%

Metrology and Measurement Systems

Organiściak,

Borkowski

2020

View full text Add to dashboard Cite

The paper examines the usage of Convolutional Bidirectional Recurrent Neural Network (CBRNN) for a problem of quality measurement in a music content. The key contribution in this approach, compared to the existing research, is that the examined model is evaluated in terms of detecting acoustic anomalies without the requirement to provide a reference (clean) signal. Since real music content may include some modes of instrumental sounds, speech and singing voice or different audio effects, it is more complex to analyze than clean speech or artificial signals, especially without a comparison to the known reference content. The presented results might be treated as a proof of concept, since some specific types of artefacts are covered in this paper (examples of quantization defect, missing sound, distortion of gain characteristics, extra noise sound). However, the described model can be easily expanded to detect other impairments or used as a pre-trained model for other transfer learning processes. To examine the model efficiency several experiments have been performed and reported in the paper. The raw audio samples were transformed into Mel-scaled spectrograms and transferred as input to the model, first independently, then along with additional features (Zero Crossing Rate, Spectral Contrast). According to the obtained results, there is a significant increase in overall accuracy (by 10.1%), if Spectral Contrast information is provided together with Mel-scaled spectrograms. The paper examines also the influence of recursive layers on effectiveness of the artefact classification task.

show abstract

“…It used training and test sets of feature vectors. In the literatures [29][30][31][32], neural networks were used for fault diagnosis [29,30], controlling a temperature eld [31], prediction of speech quality [32], and classi cation of emotion recognition [33]. e authors used three-layer backpropagation neural network for data classi cation (input layer, hidden layer, and output layer).…”

Section: Backpropagation Neural Networkmentioning

confidence: 99%

Vibration‐Based Fault Diagnosis of Commutator Motor

Głowacz

2018

Shock and Vibration

View full text Add to dashboard Cite

This paper presents a study on vibration-based fault diagnosis techniques of a commutator motor (CM). Proposed techniques used vibration signals and signal processing methods. The authors analysed recognition efficiency for 3 states of the CM: healthy CM, CM with broken tooth on sprocket, CM with broken rotor coil. Feature extraction methods called MSAF-RATIO-50-SFC (method of selection of amplitudes of frequencies ratio 50 second frequency coefficient), MSAF-RATIO-50-SFC-EXPANDED were implemented and used for an analysis. Feature vectors were obtained using MSAF-RATIO-50-SFC, MSAF-RATIO-50-SFC-EXPANDED, and sum of RSoV. Classification methods such as nearest mean (NM) classifier, linear discriminant analysis (LDA), and backpropagation neural network (BNN) were used for the analysis. A total efficiency of recognition was in the range of 79.16%–93.75% (TV). The proposed methods have practical application in industries.

show abstract

Prediction of Speech Quality Based on Resilient Backpropagation Artificial Neural Network

Cited by 8 publications

References 13 publications

Digital Transformation of Terrestrial Radio: An Analysis of Simulcasted Broadcasts in FM and DAB+ for a Smart and Successful Switchover

Digital Transformation of Terrestrial Radio: An Analysis of Simulcasted Broadcasts in FM and DAB+ for a Smart and Successful Switchover

Metrology and Measurement Systems

Vibration‐Based Fault Diagnosis of Commutator Motor

Contact Info

Product

Resources

About