Dissertation > Excellent graduate degree dissertation topics show

Single channel speech enhancement algorithm and application

Author: LuJing
Tutor: ChenJian
School: Xi'an University of Electronic Science and Technology
Course: Communication and Information System
Keywords: Single channel speech enhancement Noise Cancellation Noise spectrum estimation Auditory masking effect
CLC: TN912.35
Type: Master's thesis
Year: 2009
Downloads: 263
Quote: 1
Read: Download Dissertation

Abstract


The purpose of speech enhancement from the speech signal with noise as far as possible in extracting pure speech to improve speech naturalness and intelligibility. It is an effective solution to noise pollution, an important means to improve the signal-to-noise ratio, in recent years in the field of speech communication and speech recognition applications. This paper discusses the characteristics of speech signal and noise signal, analyzes the voice activity detection technology and noise estimation algorithm, is studied under different noise environment for a variety of speech enhancement algorithm. Based on the above discussion, this paper proposes a combination of adaptive filter based on auditory masking effect of single channel speech enhancement algorithm, combined with an ability to quickly track changes in the noise estimation algorithm, a new speech enhancement system is designed. Experiments show that the design enhance the system has the following advantages: (1) can fast track the change of the noise spectrum, can accurately estimate the noise parameters; (2) the algorithm complexity is small, good real-time; (3) can the time-varying channel and low SNR gain good strengthening effect; (4) remove the noise effect is obvious, can cover the music noise, speech intelligibility and naturalness. The single channel speech enhancement system designed by this paper has been applied to the voice noise reduction of the short-wave radio.

Related Dissertations

  1. Communication certain digital filter design,TN713.7
  2. The FECG Detection Algorithm and Implementation Based on BSS,TN911.23
  3. Line train noise canceling technology research,TN911.7
  4. Adaptive LMS algorithm and application in noise filtering,TN713
  5. Study on a Modified Spectral Subtraction Method Based on Auditory Masking Property,TN912.3
  6. Research for Harmonic Detection Method of Active Power Filter,TN713.8
  7. Ferromagnetic material separation technology research,TM271
  8. Adaptive Filter and Its Application in the gearbox feature extraction application,TN713
  9. Research on Multiple People Tracking Algorithm in Video Stream,TP391.41
  10. Research of Adaptive Noise Cancellation Technology Based on DSP,TN911.4
  11. Ultra-wideband CMOS low noise amplifier design,TN722.3
  12. Arithmetic Research on Audio Signal Restoration,TN912.3
  13. Research on Channel Estimation for MIMO-OFDM System,TN919.3
  14. The Study of Analysis and Methods of Adaptive Noise Cancellation in Communication Systems,TN911.4
  15. Research on Denoising for Radio Frequency Signals,TN92
  16. Research on Noise Cancellation under High Noise Condition,TN919.8
  17. Research on the Deep Hole Drilling Monitoring System Based on DSP,TG52
  18. Identification of T-S Fuzzy Models Using Techniques of Particle Swarm Optimization Algorithm,TP18
  19. The Study of Methods of Adaptive Noise Cancellation in Communication System,TN911.4
  20. The Research on Adaptive Speech Enhancement in Intensive Noise Envelopment,TN912.35
  21. Research of Signal Processing Technique Based General Platform in Some Radar System,TN957.51

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Electro-acoustic technology and speech signal processing > Speech Signal Processing > Speech Enhancement
© 2012 www.DissertationTopic.Net  Mobile