Dissertation > Excellent graduate degree dissertation topics show
Voice Activity Detection Based on Vehicle Embedded System
Author: LiuQiBo
Tutor: LiYanJie
School: Harbin Institute of Technology
Course: Control Science and Engineering
Keywords: Car electronic equipment Voice activity detection Cochlea equation Order statistics filtering Wiener filter Linear discriminant function
CLC: TN912.3
Type: Master's thesis
Year: 2011
Downloads: 52
Quote: 0
Read: Download Dissertation
Abstract
|
With the development of the automobile industry , the increasingly strong demand for on-board electronic equipment . In recent years the development of speech recognition technology gradually towards practical voice recognition feature embedded in the vehicle electronic equipment is becoming a hot car electronics industry . Voice signal endpoint detection is an essential part of the intelligent vehicle speech recognition device , voice endpoint detection algorithm will have a direct impact to the intelligent voice recognition accuracy and recognition speed of this article dedicated to the research vehicle speech recognition device - based speech endpoint detection algorithm to improve the accuracy of speech recognition and speech recognition speed , vehicle equipment to use more convenient and smooth . This paper presents a cochlear equation and the sub-band information fusion endpoint detection algorithm . The algorithm Wiener filtering method denoising based on the use of cochlear equation for nonlinear subband divided the spectrum of the speech signal . To estimate the sub - band signal - to - noise ratio for each sub-band , using the order statistics filtering , and finally using the linear discriminant function to integration of the SNR information of each subband , in accordance with the results in the fusion to the endpoint detection . In addition , the paper also tried based eigenvectors endpoint detection algorithm , this algorithm will be a one-dimensional voice signal into a two - dimensional image signal , the method using image processing to do the endpoint detection of speech , although because of the calculated amount of has not been adopted , but the algorithm will be a one-dimensional signal into the thinking of the two-dimensional signal processing has a positive meaning . Voice -based the 863 four major dialects database and the actual vehicle environment experimental validation algorithm able to maintain a high accuracy in the vehicle environment noise , while the complexity of the algorithm is also suitable for use in vehicle embedded environment . The successful implementation of the algorithm on the vehicle equipment significantly improves the reaction time of the vehicle equipment , voice recognition , recognition accuracy has improved to some extent , provide a strong guarantee for the practical application of vehicle equipment .
|
Related Dissertations
- Of speckle noise removal in medical ultrasound image,TP391.41
- Study of Extraction and Optimization Characteristic Parameters in Speaker Recognition,TN912.34
- Study on the Algorithms of Adaptive Optical Images Restoration,TP391.41
- The Fast Method for DOA Estimation of Wideband Signals and Its Implementation on DSP,TN911.7
- Theory and Application of Compressive Sensing Based on Wavelet Domain Wiener Filter,TN911.7
- The adaptive beam sidelobes control and nulling broadening,TN911.7
- Based on the time domain filter for speech enhancement algorithm,TN912.35
- The Research of Speech Enhancement Algorithm,TN912.35
- Based on the short-term rate spectral estimation speech enhancement method,TN912.3
- Study of Speech Enhancement and Blind Separation Algorithm with Multi-input,TN912.3
- The Research on the Voice Activity Detection in the Environment with High Noise,TN912.3
- Research on Improved Algorithm about Speech Enhancement,TN912.3
- Study of Relationship between Spatial Resolution of Ring Coded-aperture Images and the Width of Ring in ICF,TN27
- Study on Signal Acquisition & Proessing Method of the Micro Spectrometer,TH744
- Research of Shake Compensation Techniques in Image Imaging Process,TP391.41
- Research on Channel Estimation Methods in MIMO-OFDM Systems,TN919.3
- Research on Anti-noise Speech Recognition Methods Based on Robustness PLPC,TN912.34
- The Frontend Noise Reduce of Isolated Word Speech Recognition,TN912.34
- Research on Performance Improvement for ITU-T G.729,TN912.3
- Image De-noising Algorithm Via Stationary Wavelet-domain Wiener Filter Based on Correlation Model,TN713
CLC: > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Electro-acoustic technology and speech signal processing > Speech Signal Processing
© 2012 www.DissertationTopic.Net Mobile
|