Sound quality prediction of vehicle interior noise using deep belief networks

Huang, Hai; Huang, Xiao; Li, Ren X.; Lim, Teik C.; Ding, Wei P.

doi:10.1016/j.apacoust.2016.06.021

Cited by 71 publications

(25 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this experiment, the semantic differential method [26][27][28] was adopted to evaluate the directional discrimination degree of sound sources. The order scale was used for evaluation, in which directional discrimination degrees are shown as Table 4.…”

Section: Methodsmentioning

confidence: 99%

Advancement of Individualized Head-Related Transfer Functions (HRTFs) in Perceiving the Spatialization Cues: Case Study for an Integrated HRTF Individualization Method

Wang

Zeng

2019

Applied Sciences

View full text Add to dashboard Cite

Head-related transfer function (HRTF), which varies across individuals at the same direction, has grabbed widespread attention in the field of acoustics and been used in many scenarios. In order to in-depth investigate the performance of individualized HRTFs on perceiving the spatialization cues, this study presents an integrated algorithm to obtain individualized HRTFs, and explores the advancement of such individualized HRTFs in perceiving the spatialization cues through two different binaural experiments. An integrated method for HRTF individualization on the use of Principle Component Analysis (PCA), Multiple Linear Regression (MLR) and Partial Least Square Regression (PLSR) was presented first. The objective evaluation was then made to verify the algorithmic effectiveness of that method. Next, two subjective experiments were conducted to explore the advancement of individualized HRTFs in perceiving the spatialization cues. One was auditory directional discrimination degree based on semantic differential method, in which the azimuth information of sound sources was told to the listeners before listening. The other was auditory localization, in which the azimuth information was not told to the listeners before listening. The corresponding statistical analyses for the subjective experimental results were made. All the experimental results support that individualized HRTFs obtained from the presented method achieve a preferable performance in perceiving the spatialization cues.

show abstract

Section: Methodsmentioning

confidence: 99%

Advancement of Individualized Head-Related Transfer Functions (HRTFs) in Perceiving the Spatialization Cues: Case Study for an Integrated HRTF Individualization Method

Wang

Zeng

2019

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…To classify sound patterns, some techniques based on the artificial neural network (ANN) and the support vector machine (SVM) methods (virtual "brain") were used for predicting the loudness, sharpness and annoyance indices of vehicle noise. [23][24][25] Compared with the loudness and sharpness indices, the roughness of a sound is more difficult in the feature extraction and perception modelling, because there exists hardly any correlation between the subjective impressions of test persons and available roughness parameters. 17 To recognize the patterns of auditory roughness, the ANN used in the loudness and sharpness predictions needs to be reconstructed and modified.…”

Section: Introductionmentioning

confidence: 99%

Roughness Evaluation Approach for Nonstationary Vehicle Noise Based on Wavelet Packet and Neural Network Techniques

Guo¹,

Wang²,

Wang³

et al. 2018

IJAV

View full text Add to dashboard Cite

Based on the wavelet packet decomposition (WPD) and artificial neural network (ANN) methods, this paper presents a new technique for auditory roughness evaluation (ARE) of nonstationary vehicle noise, named WPD-ANN-ARE model. According to sound transfer and perception by the human auditory system, the noise roughness of a sample vehicle under different conditions of constant speed, acceleration, and braking are calculated. After comparisons by the time-frequency analysis techniques in common use, a WPD model with approximately twentyone critical bands, which is specially designed by considering the auditory perception characteristics of human, is proposed for envelope feature extraction of vehicle noise. Taking the WPD extracted features as inputs and the calculated roughnesses as outputs, a back-propagation ANN with one hidden layer is trained and established for ARE of nonstationary vehicle noises. The verification results show that errors of the time-varying roughness calculated from the WPD-ANN-ARE are below 8.56 percent, which suggest a very good accuracy of the newly proposed ARE model. In applications, the WPD-ANN-ARE can be directly used in ARE of vehicle noises. And the modelling approach presented in this paper may be extended to other sound related fields for sound quality evaluation (SQE) in engineering.

show abstract

“…Moreover, deep belief networks (DBNs) are used to train each subnetwork in modular neural network for the following reasons. Firstly, the DBNs have been successfully applied to traffic flow prediction and sound quality prediction . Analysis of the similarity between QoE/QoS correlation modeling and the above two prediction problems has motivated us to use DBN to address the QoE prediction for satellite multimedia services.…”

Section: Introductionmentioning

confidence: 99%

“…Firstly, the DBNs have been successfully applied to traffic flow prediction 29 and sound quality prediction. 30 Analysis of the similarity between QoE/QoS correlation modeling and the above two prediction problems has motivated us to use DBN to address the QoE prediction for satellite multimedia services. Secondly, although traditional neural networks have been used in QoE/QoS mapping for multimedia services in terrestrial networks, the results show that deeper and richer architectures are required to solve this issue due to the limited capability of traditional neural networks.…”

mentioning

confidence: 99%

Modular and deep QoE/QoS mapping for multimedia services over satellite networks

Wang

Huang

2018

Int J Communication

View full text Add to dashboard Cite

Summary Satellite networks are promising alternatives to deal with the increasing traffic volume and provide universal access for end users. Quality of experience (QoE) as an aggregate of user perception and network quality will be a critical success factor for multimedia service delivery over satellite networks. Quality of experience prediction based on quality of service (QoS) has received much attention recently, but it has been little investigated in satellite networks. In this paper, on the basis of the modular neural network and deep belief networks (DBNs), we design a QoE/QoS mapping method for multimedia services over satellite networks to translate QoS parameters into user QoE. The task of QoE/QoS mapping for multimedia services is decomposed into several different subtasks by using traffic classification. Heterogeneous DBNs are used to learn the mapping relationships between QoE and QoS for different types of services (ie, subtask) simultaneously. An integrative approach based on relative distance is exploited to select subneural network(s) to predict resulting QoE collaboratively. To determine the weight parameters of the QoE/QoS correlation model, the dataset composed of QoS parameters and subjective opinion scores is built on the basis of satellite network simulator and subjective test. Finally, the effectiveness of our QoE/QoS mapping method is validated, and the relationship between adoption rating and opinion scores is analyzed.

show abstract

Sound quality prediction of vehicle interior noise using deep belief networks

Cited by 71 publications

References 46 publications

Advancement of Individualized Head-Related Transfer Functions (HRTFs) in Perceiving the Spatialization Cues: Case Study for an Integrated HRTF Individualization Method

Advancement of Individualized Head-Related Transfer Functions (HRTFs) in Perceiving the Spatialization Cues: Case Study for an Integrated HRTF Individualization Method

Roughness Evaluation Approach for Nonstationary Vehicle Noise Based on Wavelet Packet and Neural Network Techniques

Modular and deep QoE/QoS mapping for multimedia services over satellite networks

Contact Info

Product

Resources

About