A Speech Quality Classifier based on Signal Information that Considers Wired and Wireless Degradations

Militani, Davi Ribeiro; Begazo, Dante Coaquira; Rosa, Renata Lopes; Rodríguez, Demóstenes Zegarra

doi:10.23919/softcom.2019.8903776

Cited by 1 publication

(8 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The RBM can model fragments of a signal [54] and the RBM can also be used for supervised techniques [26]. The supervised learning was proposed by the DRBM algorithm, in which labels or classes information are incorporate into the visible layer (input); thus, the joint distribution of the input data are calculated and they are classified in a corresponding label.…”

Section: Speech Characterization and Classification Modelsmentioning

confidence: 99%

“…RBM is a generative stochastic Artificial Neural Network (ANN) that can learn a probability distribution; for classifications purposes, it is necessary to add a supervised learning method, classifying the samples based on the characteristics extracted by the RBM. Studies regarding the characteristics identification in speech signals demonstrate superior accuracy of the RBM in relation to other widely used methods [25], [26], such as Support Vector Machine (SVM). In a previous work [26], a non-intrusive speech quality classifier based on Discriminative Restricted Boltzmann Machines (DRBM) is presented, and it reached a high accuracy for classifying a MOS speech quality.…”

Section: Introductionmentioning

confidence: 99%

“…Studies regarding the characteristics identification in speech signals demonstrate superior accuracy of the RBM in relation to other widely used methods [25], [26], such as Support Vector Machine (SVM). In a previous work [26], a non-intrusive speech quality classifier based on Discriminative Restricted Boltzmann Machines (DRBM) is presented, and it reached a high accuracy for classifying a MOS speech quality. However, the DRBM presents some deficiencies in terms of the training time, and some proposals try to decrease the impact of this problem [27]- [29].…”

Section: Introductionmentioning

confidence: 99%

“…In this context, this research presents additional contributions regarding to our previous work [26], which can be stated as follows:…”

Section: Introductionmentioning

confidence: 99%

“…• [21], [26], [35]. To determine the proposed model, different speech signal characteristics are used.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Speech Quality Classifier based on Tree-CNN Algorithm that Considers Network Degradations

Vieira

Rosa

Rodríguez

2020

J. commun. softw. syst. (Online)

View full text Add to dashboard Cite

Many factors can affect the users’ quality of experience (QoE) in speech communication services. The impairment factors appear due to physical phenomena that occur in the transmission channel of wireless and wired networks. The monitoring of users’ QoE is important for service providers. In this context, a non-intrusive speech quality classifier based on the Tree Convolutional Neural Network (Tree-CNN) is proposed. The Tree-CNN is an adaptive network structure composed of hierarchical CNNs models, and its main advantage is to decrease the training time that is very relevant on speech quality assessment methods. In the training phase of the proposed classifier model, impaired speech signals caused by wired and wireless network degradation are used as input. Also, in the network scenario, different modulation schemes and channel degradation intensities, such as packet loss rate, signal-to-noise ratio, and maximum Doppler shift frequencies are implemented. Experimental results demonstrated that the proposed model achieves significant reduction of training time, reaching 25% of reduction in relation to another implementation based on DRBM. The accuracy reached by the Tree-CNN model is almost 95% for each quality class. Performance assessment results show that the proposed classifier based on the Tree-CNN overcomes both the current standardized algorithm described in ITU-T Rec. P.563 and the speech quality assessment method called ViSQOL.

show abstract

Section: Speech Characterization and Classification Modelsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…In this context, this research presents additional contributions regarding to our previous work [26], which can be stated as follows:…”

Section: Introductionmentioning

confidence: 99%

“…• [21], [26], [35]. To determine the proposed model, different speech signal characteristics are used.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Speech Quality Classifier based on Tree-CNN Algorithm that Considers Network Degradations

Vieira

Rosa

Rodríguez

2020

J. commun. softw. syst. (Online)

View full text Add to dashboard Cite

show abstract

A Speech Quality Classifier based on Signal Information that Considers Wired and Wireless Degradations

Cited by 1 publication

References 31 publications

A Speech Quality Classifier based on Tree-CNN Algorithm that Considers Network Degradations

A Speech Quality Classifier based on Tree-CNN Algorithm that Considers Network Degradations

Contact Info

Product

Resources

About