A New Method Based on HMMs and K-Means Algorithms for Noise-Robust Voice Activity Detector

2011 ◽  
Vol 128-129 ◽  
pp. 461-464 ◽  
Author(s):  
Bing Luo ◽  
Zheng Pei ◽  
Li Xu ◽  
Da Li Hu
Keyword(s):  
Markov Models ◽  
Signal To Noise Ratio ◽  
Adaptive Threshold ◽  
High Signal ◽  
Ratio Test ◽  
Signal To Noise ◽  
Voice Activity Detector ◽  
Noise Rate ◽  
Noise Robust ◽  
Voice Activity

In this paper, we proposed left-right hidden Markov models (HMMs) combination with k-means threshold of Likelihood ratio test (LRT) to identify the start and end of the speech. This method builds two models of non-speech and speech but not two states, i.e. each model could conclude several states. In the experiments we present the Voice Activity Detection (VAD) results between two states hidden semi-Markov model (HSMM) and proposed algorithm. We also compare accuracy and robust between the k-means threshold and the adaptive threshold in high signal to noise rate in the background noise. It presents that k-means threshold is more effective than the adaptive threshold and the proposed method also make a better performance than two states HSMM based VAD, especially in the low signal-to-noise ratio (SNR) environment.

Noise-Robust Voice Activity Detector Based on Four States-Based HMM

2013 ◽  
Vol 411-414 ◽  
pp. 743-748 ◽  
Author(s):  
Bin Zhou ◽  
Jing Liu ◽  
Zheng Pei
Keyword(s):  
Fractal Dimension ◽  
Hidden Markov ◽  
Signal To Noise Ratio ◽  
Activity Detection ◽  
Noisy Environments ◽  
Signal To Noise ◽  
Voice Activity Detector ◽  
Noise Robust ◽  
Voice Activity ◽  
Accuracy Performance

Voice activity detection (VAD) is more and more essential in the noisy environments to provide an accuracy performance in the speech recognition. In this paper, we provide a method based on left-right hidden Markov model (HMM) to identify the start and end of the speech. The method builds two models of non-speech and speech instead of existed two states, formally, each model could include several states, we also analysis other features, such as pitch index, pitch magnitude and fractal dimension of speech and non-speech.. We compare the VAD results with the proposed algorithm and two states HMM. Experiments show that the proposed method make a better performance than two states HMMs in VAD, especially in the low signal-to-noise ratio (SNR) environment.

scholarly journals Adaptive Threshold-Aided K-Best Sphere Decoding for Large MIMO Systems

2019 ◽  
Vol 9 (21) ◽  
pp. 4624
Author(s):  
Uzokboy Ummatov ◽  
Kyungchun Lee
Keyword(s):  
Mimo Systems ◽  
Signal To Noise Ratio ◽  
Multiple Input Multiple Output ◽  
Threshold Value ◽  
Adaptive Threshold ◽  
Sphere Decoding ◽  
High Signal ◽  
Signal To Noise ◽  
Multiple Input ◽  
Input Multiple Output

This paper proposes an adaptive threshold-aided K-best sphere decoding (AKSD) algorithm for large multiple-input multiple-output systems. In the proposed scheme, to reduce the average number of visited nodes compared to the conventional K-best sphere decoding (KSD), the threshold for retaining the nodes is adaptively determined at each layer of the tree. Specifically, we calculate the adaptive threshold based on the signal-to-noise ratio and index of the layer. The ratio between the first and second smallest accumulated path metrics at each layer is also exploited to determine the threshold value. In each layer, in addition to the K paths associated with the smallest path metrics, we also retain the paths whose path metrics are within the threshold from the Kth smallest path metric. The simulation results show that the proposed AKSD provides nearly the same bit error rate performance as the conventional KSD scheme while achieving a significant reduction in the average number of visited nodes, especially at high signal-to-noise ratios.

A Fast Adaptive Beamforming Algorithm Based on Gram-Schmidt Orthogonalization

2021 ◽  
Vol 16 (4) ◽  
pp. 642-650
Author(s):  
Bing-Feng Qian ◽  
Shi-Jie Gao ◽  
Quan-Feng Li ◽  
Qian Zhang ◽  
Ye Wang
Keyword(s):  
Computational Complexity ◽  
Covariance Matrix ◽  
Signal To Noise Ratio ◽  
Data Preprocessing ◽  
Adaptive Threshold ◽  
Adaptive Beamforming ◽  
High Signal ◽  
Signal To Noise ◽  
Schmidt Orthogonalization ◽  
Beamforming Algorithm

Fast implementation is one of the important indexes of the ADBF algorithm. The advantages of the Gram-Schmidt (GS) orthogonalization algorithm are that it can reconstruct the interference subspace well under the high signal-to-noise ratio and has fast convergence speed and low computational complexity. This paper studies the RGS algorithm for GS orthogonalization of sampling covariance matrix. To estimate the interference subspace more accurately, this paper modifies the orthogonal adaptive threshold of covariance matrix, and extends the proposed GS orthogonal algorithm of covariance matrix based on data preprocessing to the adaptive beamforming processing at subarray level.

An improved noise-robust voice activity detector based on hidden semi-Markov models

2011 ◽  
Vol 32 (7) ◽  
pp. 1044-1053 ◽  
Author(s):  
Yuan Liang ◽  
Xianglong Liu ◽  
Yihua Lou ◽  
Baosong Shan
Keyword(s):  
Markov Models ◽  
Voice Activity Detector ◽  
Noise Robust ◽  
Voice Activity

A compact guided-wave EMAT with pulsed electromagnet for ferromagnetic tube inspection

2020 ◽  
Vol 64 (1-4) ◽  
pp. 951-958
Author(s):  
Tianhao Liu ◽  
Yu Jin ◽  
Cuixiang Pei ◽  
Jie Han ◽  
Zhenmao Chen
Keyword(s):  
Power Plants ◽  
Signal To Noise Ratio ◽  
Small Diameter ◽  
Performance Testing ◽  
Guided Wave ◽  
High Signal ◽  
Receiver Coil ◽  
Signal To Noise ◽  
Corrosion Defects

Small-diameter tubes that are widely used in petroleum industries and power plants experience corrosion during long-term services. In this paper, a compact inserted guided-wave EMAT with a pulsed electromagnet is proposed for small-diameter tube inspection. The proposed transducer is noncontact, compact with high signal-to-noise ratio and unattractive to ferromagnetic tubes. The proposed EMAT is designed with coils-only configuration, which consists of a pulsed electromagnet and a meander pulser/receiver coil. Both the numerical simulation and experimental results validate its feasibility on generating and receiving L(0,2) mode guided wave. The parameters for driving the proposed EMAT are optimized by performance testing. Finally, feasibility on quantification evaluation for corrosion defects was verified by experiments.

Narrow-frequency sharp-angular filters using all-dielectric cascaded meta-gratings

Nanophotonics ◽  
2020 ◽  
Vol 9 (10) ◽  
pp. 3443-3450 ◽  
Author(s):  
Wei-Nan Liu ◽  
Rui Chen ◽  
Wei-Yi Shi ◽  
Ke-Bo Zeng ◽  
Fu-Li Zhao ◽  
...  
Keyword(s):  
Thin Film ◽  
Signal To Noise Ratio ◽  
Incident Angle ◽  
Design Strategy ◽  
Transmission Peak ◽  
High Signal ◽  
Waveguide Array ◽  
Signal To Noise ◽  
Center Wavelength ◽  
Noise Ratio

AbstractSelective transmission or filtering always responds to either frequency or incident angle, so as hardly to maximize signal-to-noise ratio in communication, detection and sensing. Here, we propose compact meta-filters of narrow-frequency sharp-angular transmission peak along with broad omnidirectional reflection sidebands, in all-dielectric cascaded subwavelength meta-gratings. The inherent collective resonance of waveguide-array modes and thin film approximation of meta-grating are employed as the design strategy. A unity transmission peak, locating at the incident angle of 44.4° and the center wavelength of 1550 nm, is demonstrated in a silicon meta-filter consisting of two-layer silicon rectangular meta-grating. These findings provide possibilities in cascaded meta-gratings spectroscopic design and alternative utilities for high signal-to-noise ratio applications in focus-free spatial filtering and anti-noise systems in telecommunications.

scholarly journals Low-dose, phase-contrast mammography with high signal-to-noise ratio

2016 ◽  
Vol 7 (2) ◽  
pp. 381 ◽  
Author(s):  
Lukas B. Gromann ◽  
Dirk Bequé ◽  
Kai Scherer ◽  
Konstantin Willer ◽  
Lorenz Birnbacher ◽  
...  
Keyword(s):  
Phase Contrast ◽  
Low Dose ◽  
Signal To Noise Ratio ◽  
High Signal ◽  
Signal To Noise ◽  
Noise Ratio
Sign in / Sign up

Export Citation Format

Share Document