Multi-band frequency compression for improving speech perception by listeners with moderate sensorineural hearing loss

Speech enhancement and multi-band frequency compression for suppression of noise and intraspeech spectral masking in hearing aids

“…Use of fixed frame analysissynthesis resulted in perceptible distortions and adversely affected the advantage of multi-band frequency compression. Evaluation of pitch-synchronous implementation using Modified Rhyme Test on eight subjects with moderate sensorineural loss showed best performance for the compression factor of 0.6, with a mean increase of 16.5% in recognition scores and a mean decrease of 0.89 s in response time [9]. The results indicated that the processing may improve the usefulness of hearing aids for persons with moderate sensorineural loss.…”

“…Multi-band frequency compression [7] - [9] involves dividing speech spectrum into analysis bands and compressing the spectral samples in each band towards the band center. The objective is to present speech energy in relatively narrow bands in order to avoid masking by adjacent spectral components and to reduce the effects of increased intraspeech spectral masking.…”

“…Arai et al [7] applied the technique using auditory critical bandwidths on the magnitude spectrum and the complex spectrum was obtained by associating the compressed magnitude spectrum with the original phase spectrum. For decreasing the computation and reducing the processing related artifact, Kulkarni et al [9] applied the compression on the complex spectrum without calculating the magnitude and phase spectra. Investigations with different types of bandwidths, frequency mappings for spectral modification, and segmentation for analysis-synthesis showed that maximum improvement in speech perception was obtained for auditory critical bandwidth based compression using spectral segment mapping and pitchsynchronous analysis-synthesis.…”

“…In addition to providing frequency-selective gain and automatic gain control, many hearing aids have multi-channel dynamic range compression with settable attack time, release time, number of channels, and compression ratios [3], [4]. Spectral contrast enhancement [5], [6] and multi-band frequency compression [7] - [9] have been used for reducing the effect of increased intraspeech spectral masking caused by widening of auditory filters. Listeners with sensorineural loss experience severe difficulty in noisy environments.…”

A Thorough Investigation on Speech Enhancement Techniques for Hearing Aids

“…In the past decades, the binaural dichotic presentation [2], spectral contrast enhancement [3], and multiband frequency compression [4] algorithm were proposed for reducing the spectral masking but these techniques were observed to have problems with speech perception. After this, the redundant signal energy is minimized in speech segregation algorithm for attaining the fine quality of a desired speech signal.…”