Noise reduction algorithms for digital hearing instruments
Volkmar Hamacher, Inga Holube · The Journal of the Acoustical Society of America · 1999
One of the main problems for hearing instrument users is the reduction of speech intelligibility in noisy and/or reverberant environment. To compensate for this disadvantage, noise reduction algorithms preceding or integrated in the hearing instruments’ signal processing are required. New generations of powerful digital hearing instruments enable the implementation of sophisticated algorithms, which have been partly applied in other communication devices. These algorithms offered promising benefit in previous laboratory studies with hearing instrument users. This paper presents a survey of noise reduction strategies implemented in current hearing instruments and new algorithms applicable in future devices will be given. The algorithms can be divided into different classes concerning their approach in noise reduction. One class of algorithms uses the signal from one microphone and reduces the noise by analysis of the signal characteristics. Another class of noise reduction techniques uses several microphones and exploits spatial information about the sound field to improve the signal-to-noise ratio. Restrictions for the different algorithms are the computational effort and the preconditions to the interfering noise field. The expected benefit for hearing instrument users in typical everyday life hearing situations will be discussed.