Adaptive noise suppression in voice communication using a neuro-fuzzy inference system

This paper describes the implementation of a combination of techniques of the fuzzy system and artificial intelligence in the application area of non-linear suppression of noise and interference. The structure used is called ANFIS (Adaptive Neuro Fuzzy Inference System). This system finds practical use mainly in audio telephone (mobile) communication in a noisy environment (transport, production halls, sports matches, etc.). Within the experiments carried out, the authors created, based on the ANFIS structure, a comprehensive system for the adaptive suppression of unwanted background interference that occurs in audio communication and which degrades the audio signal. The system designed has been tested on real voice signals. Noise cancellation performance of the algorithms has been compared by means of SSNR (Segmental Signal to Noise Ratio) and DTW (Dynamic Time Warping). Also processing durations of the algorithms are determined for evaluating the possibility of real time implementation. The results imply that a system using ANFIS has better experimental results than conventional systems built on adaptive algorithms of the LMS (Least Mean Squares) and RLS (Recursive Least Squares) families.