Dissertation > Excellent graduate degree dissertation topics show

Research on Speech Enhancement Algorithms for Digital Hearing Aids

Author: GaoJie
Tutor: HuGuangShu
School: Tsinghua University
Course: Biomedical Engineering
Keywords: Digital hearing aids Speech Enhancement Wiener filter Beamforming
CLC: R764.5
Type: Master's thesis
Year: 2005
Downloads: 351
Quote: 4
Read: Download Dissertation

Abstract


With continuous aging of society and people concerned about the problem of deafness, hearing aids development of growing attention has been paid. In recent years, digital hearing aids, with its small size, low power consumption, strong adjustable, becoming the protagonist of the hearing aid market. The high-speed, low power consumption, the introduction of multi-function digital signal processor, and also provides a good platform for digital hearing aids algorithm. In this thesis, the main work of digital hearing aids speech enhancement algorithm, combined with the characteristics of the digital signal processor structure development for digital hearing aids, efficient, real-time speech enhancement algorithm. The thesis designs one for single-channel digital hearing aids, speech enhancement algorithm implemented in real time, and based on TI's TMS320VC5502 digital signal processor-based platforms. Through the comparison of various algorithms, the final selection of a voice activity detection algorithm based on energy envelope identify with voice and non-voice area, from the noisy speech signal, and then take advantage of the free speech zone noise statistical properties through The power spectrum estimation, Wiener filtering method to enhance the voice. Residual musical noise enhancement processing, smoothing and average power spectrum to be suppressed. The overall algorithm is compact, effective and easy to be implemented on a digital signal processor and optimize. Defects for single channel speech enhancement, further applications of digital hearing aids microphone array speech enhancement beam forming algorithm, focusing on analysis of the ultra-direction and adaptive beamforming algorithm and algorithm parameters on the beam formed Effect be analyzed. In addition, for the defects of traditional adaptive beam forming algorithm, this paper proposed algorithm considering noise suppression effect and the distortion of the target speech signal, modifying the cost function of the adaptive filter, to redefine the the NLMS algorithm's iterative formula to reduce self- adapt to the loss of the voice in the process. The simulation results show that using the new cost function, background noise and strong interference has been inhibited, the target voice signal distortion is also significantly reduced, thus obtained to improve the robustness of the algorithm to work in the actual environment. And the vast majority of the algorithm operation are easy to utilize the characteristics of the hardware structure of the digital signal processor, efficient implementation. Single-channel digital hearing aids speech enhancement algorithms and the application of the microphone array speech enhancement beamforming algorithm, respectively, according to the different characteristics of the noise for noise suppression, has a certain degree of complementarity between the two digital hearing aids can be used in combination to better speech enhancement effects.

Related Dissertations

  1. The Technology Research on Channel Coding and Channel Estimation in MC-CDMA Systems,TN929.533
  2. Research on Chinese Speech Processing and Speech Enhancement in Hearing Aids,TN912.3
  3. A Beamforming Algorithm Based Relay Matrix,TN929.5
  4. Study of Interferencealignment in Coordinated Multiple Points System,TN92
  5. Of speckle noise removal in medical ultrasound image,TP391.41
  6. Research and Implementation of Algorithm for Restoration of Motion Blurred Images,TP391.41
  7. Research of Underwater Image Restoration Algorithm Based on Backward Scattering Noise Model,TP391.41
  8. Common types blurred image restoration Research and Implementation,TP391.41
  9. Research on MIMO Downlink Beamforming,TN919.3
  10. Study on the Algorithms of Adaptive Optical Images Restoration,TP391.41
  11. The Fast Method for DOA Estimation of Wideband Signals and Its Implementation on DSP,TN911.7
  12. Passive Source Localization of Fiber-optic Hydrophone Array by Focused Beamforming,U666.7
  13. Application of Joint Beamforming and STC in Multi-Antenna Systems,TN92
  14. Wideband digital array real-time delay technology,TN911.7
  15. Real- array control of the data processing system implemented with the beam forming,TN911.7
  16. Radiation source passive radar signal processing simulation,TN958.97
  17. Based on LFM UWB localization,TN95
  18. Beamforming based noise source identification and visualization of the sound field,TB534.1
  19. Simulation Research of Imaging Algorithm for 3-D Imaging Sonar,U666.7
  20. The Research and Improvement of Smart Antenna in TD-SCDMA System,TN929.533
  21. Microphone Array Speech Enhancement Based on Filterbanks Generalized Sidelobe Canceller,TN641

CLC: > Medicine, health > Otorhinolaryngology > Otology,ear disease > Pathology audiology and hearing aids
© 2012 www.DissertationTopic.Net  Mobile