Title :
Neural network based adaptive microphone array system for speech enhancement
Author :
Grbic, Nedelko ; Dahl, Mattias ; Claesson, Ingvar
Author_Institution :
Inst. of Signal Process., Karlsruhe Univ., Germany
Abstract :
Presents a microphone array system for use in handsfree mobile telephone equipment. The array is based on a fast and efficient “on-site” and “self-calibration” scheme. The performance in suppressing the interior car cabin noise and the far-end speech is approximately 17 dB, respectively, while maintaining the near-end speaker level. The near-end signal is almost undistorted. The performance of two different algorithms, normalized least-mean-square (NLMS) and fully connected backpropagation supervised neural network (MLP-NN) are evaluated. The proposed microphone array calibration scheme can also be used in other situations such as speech recognition devices
Keywords :
acoustic transducer arrays; adaptive filters; backpropagation; calibration; echo suppression; land mobile radio; least mean squares methods; microphones; multilayer perceptrons; radio equipment; speech enhancement; telecommunication computing; 17 dB; MLP-NN; NLMS; far-end speech; fully connected backpropagation supervised neural network; handsfree mobile telephone equipment; interior car cabin noise; microphone array calibration scheme; near-end signal; near-end speaker level; neural network based adaptive microphone array system; normalized least-mean-square; performance; self-calibration; speech enhancement; speech recognition devices; Adaptive arrays; Adaptive systems; Backpropagation algorithms; Calibration; Microphone arrays; Neural networks; Noise level; Speech enhancement; Speech recognition; Telephone equipment;
Conference_Titel :
Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on
Conference_Location :
Anchorage, AK
Print_ISBN :
0-7803-4859-1
DOI :
10.1109/IJCNN.1998.687198
