Estudio comparativo de técnicas digitales de reducción de ruido para su posible uso en auxiliares auditivos
In this paper we present a comparative study of four signal processing techniques focused to solve the environmental noise problem in hearing aids, aiming to restore speech communication to the maximum possible degree. The evaluated techniques are the RACE adaptive filter, the Spectral Subtraction, the Singular Value Decomposition based noise reduction and the Independent Component Analysis. The techniques were evaluated by computing the improvement to the signal-to-noise ratio in different environments and performing statistical significance tests on the results. The Independent Component Analysis obtained the best results, which in a simulation environment is able to restore the speech signal even when its energy level is 15dB under that of the noise.
Once an article is published in the RMIB the authors give the copyright of the published work to the RMIB, and it is authorized to publish this work in any medium, without any limitations and also, without any cost. The authors may reuse portions of the work in other documents, as well as reproduce it partially or totally for their personal use, as long as a bibliographic reference is made to the RMIB and a copy of the writing is sent.