Journal ArticleDOI
Nonintrusive Quality Assessment of Noise Suppressed Speech With Mel-Filtered Energies and Support Vector Regression
Reads0
Chats0
TLDR
This paper proposes a nonintrusive metric for the quality assessment of noise-suppressed speech and utilizes the sensitivity of FBEs to noise in order to obtain an effective representation of speech towards quality assessment.Abstract:
Objective speech quality assessment is a challenging task which aims to emulate human judgment in the complex and time consuming task of subjective assessment. It is difficult to perform in line with the human perception due the complex and nonlinear nature of the human auditory system. The challenge lies in representing speech signals using appropriate features and subsequently mapping these features into a quality score. This paper proposes a nonintrusive metric for the quality assessment of noise-suppressed speech. The originality of the proposed approach lies primarily in the use of Mel filter bank energies (FBEs) as features and the use of support vector regression (SVR) for feature mapping. We utilize the sensitivity of FBEs to noise in order to obtain an effective representation of speech towards quality assessment. In addition, the use of SVR exploits the advantages of kernels which allow the regression algorithm to learn complex data patterns via nonlinear transformation for an effective and generalized mapping of features into the quality score. Extensive experiments conducted using two third party databases with different noise-suppressed speech signals show the effectiveness of the proposed approach.read more
Citations
More filters
Journal ArticleDOI
Mulsemedia: State of the Art, Perspectives, and Challenges
TL;DR: A historic perspective on mulsemedia work is presented and current developments in the area are reviewed and standardization efforts, via the MPEG-V standard, are described.
Proceedings ArticleDOI
Quality-Net: An End-to-End Non-intrusive Speech Quality Assessment Model Based on BLSTM.
TL;DR: In this paper, an end-to-end, non-intrusive speech quality evaluation model, termed Quality-Net, based on bidirectional long short-term memory (LSTM) was proposed.
Posted Content
Quality-Net: An End-to-End Non-intrusive Speech Quality Assessment Model based on BLSTM
TL;DR: In this article, an end-to-end, non-intrusive speech quality evaluation model, termed Quality-Net, based on bidirectional long short-term memory (LSTM) was proposed.
Journal ArticleDOI
Long-Term Spectral Statistics for Voice Presentation Attack Detection
TL;DR: Investigations on ASVspoof 2015 challenge database and AVspoof database show that the proposed approach with a linear discriminative classifier yields a better system, irrespective of whether the spoofed signal is replayed to the microphone or is directly injected into the system software process.
Proceedings ArticleDOI
Novel deep autoencoder features for non-intrusive speech quality assessment
Meet H. Soni,Hemant A. Patil +1 more
TL;DR: Quantification of the experimental results suggests that proposed metric gives more accurate and correlated scores than an existing benchmark for objective, non-intrusive quality assessment metric ITU-T P.563 standard.
References
More filters
Journal ArticleDOI
Low-Complexity, Nonintrusive Speech Quality Assessment
TL;DR: A low-complexity algorithm for monitoring the speech quality over a network that can be computed from commonly used speech-coding parameters without explicit distortion modeling is described.
Journal ArticleDOI
ANIQUE: an auditory model for single-ended speech quality estimation
TL;DR: The proposed auditory non-intrusive quality estimation (ANIQUE) model is based on the functional roles of human auditory systems and the characteristics of human articulation systems and demonstrates the effectiveness of the proposed model.
Proceedings ArticleDOI
Detection of abrupt spectral changes using support vector machines an application to audio signal segmentation
Manuel Davy,Simon J. Godsill +1 more
TL;DR: An hybrid time-frequency/support vector machine algorithm for the detection of abrupt spectral changes using a stationarity index derived from support vector novelty detection theory by using sub-images extracted from the time- frequencies as feature vectors is introduced.
Journal ArticleDOI
Single-Ended Speech Quality Measurement Using Machine Learning Methods
Tiago H. Falk,Wai-Yip Chan +1 more
TL;DR: A novel single-ended algorithm constructed from models of speech signals, including clean and degraded speech, and speech corrupted by multiplicative noise and temporal discontinuities, found to be more effective than P.563, the current "state-of-art" standard single- ended algorithm.
Journal ArticleDOI
Speaker Verification Using Support Vector Machines and High-Level Features
TL;DR: A method of speaker modeling based upon support vector machines based upon linearizing a log likelihood ratio scoring system is described and generalizations of this method are shown to produce excellent results on a variety of high-level features.
Related Papers (5)
Single-Ended Speech Quality Measurement Using Machine Learning Methods
Tiago H. Falk,Wai-Yip Chan +1 more