科研之友
微信
新浪微博
Facebook
分享链接
摘要
Data-driven computing and using data for strategic advantages are exemplified by communication systems, and the speech intelligibility in communication systems is generally interrupted by interfering noise. This interference comes from the environmental noise, so it can reduce them intelligibility by masking the interested signal. An important work in communication systems is to extract speech from noisy speech and inhibiting background noise. The primary purpose of speech noise reduction system is to extract pure speech from speech signal with noise. The focus of this paper is to build a new noise reduction system on the basis of the optimization of digital noise reduction algorithms. According to the program simulation results based on MATLAB, the digital noise reduction system has many improved performances in the low SNR and achieves more than 5dB-15dB on noise reduction. The combined algorithm was tested under different noise conditions, and data display that the optimize performance of algorithm achieve the best. The simulation results demonstrate that it can get nearly three times better than the other two algorithms. The output signal of combined algorithm are very close to the pure speech signal, the performance of restore the voice signal is better than the other two algorithms.
- 出版日期2016-9
- 单位上海理工大学
