Adaptive spectral contrast enhancement based on masking effect for the hearing impaired

A spectral enhancement technique has been developed to compensate for the reduced masking effect in people with sensorineural hearing impairment. The speech is modified with the goal of increasing asymmetrically the amplitudes of the first three formant peaks relative to surrounding spectrum levels. The shape and degree of the spectral contrast alteration can be flexibly adjusted in this proposed scheme based on the reduced masking effect. The processed speech was presented to listeners with moderate to severe sensorineural hearing loss. Experiment results have shown that most of listeners consistently preferred the processed speech to the original one. A comparison of the results with Bunnell's (1990) scheme are also reported, which demonstrates furthermore the effectiveness and correctness of our scheme.