Abstract
In hearing aid devices, speech enhancement techniques are a critical component to enable users with hearing loss to attain improved speech quality under noisy conditions. Recently, the deep denoising autoencoder (DDAE) was adopted successfully for recovering the desired speech from noisy observations. However, a single DDAE cannot extract contextual information sufficiently due to the poor generalization in an unknown signal-to-noise ratio (SNR), the local minima, and the fact that the enhanced output shows some residual noise and some level of discontinuity. In this paper, we propose a hybrid approach for hearing aid applications based on two stages: (1) the Wiener filter, which attenuates the noise component and generates a clean speech signal; (2) a composite of three DDAEs with different window lengths, each of which is specialized for a specific enhancement task. Two typical high-frequency hearing loss audiograms were used to test the performance of the approach: Audiogram 1 = (0, 0, 0, 60, 80, 90) and Audiogram 2 = (0, 15, 30, 60, 80, 85). The hearing-aid speech perception index, the hearing-aid speech quality index, and the perceptual evaluation of speech quality were used to evaluate the performance. The experimental results show that the proposed method achieved significantly better results compared with the Wiener filter or a single deep denoising autoencoder alone.
Highlights
10% of the population suffers from some degree of hearing loss (HL) (Figure 1)due to overexposure to noise—both long-term, repeated exposure to noise and one-time exposure to a powerful sound that causes damage to the auditory system [1]
The results indicate that HC-deep denoising autoencoder (DDAE) provided significantly higher Hearing Aid Speech Perception Index (HASPI) scores than the other methods (i.e., Wiener filter and single DDAE separately) under most of the tested conditions
Two kinds of experiments were presented in this paper, in the first experiment; five types of noises were used and were added to the training set by corrupting random training speech signals at four signal-to-noise ratio (SNR) levels (i.e., 0, 5, 10, 15 dB)
Summary
10% of the population suffers from some degree of hearing loss (HL) (Figure 1). Organization (WHO), in 2018 [2,3], roughly 466 million people had HL.\, and it is estimated that over 900 million people will have hearing loss by 2050. The usage of hearing aids and amplifying devices is the most common treatment method. Only a small percentage of potential wearers use a hearing aid due to the general problem of enhancing speech in a noisy environment. One of the major complaints from hearing aid wearers involves the devices’ lack of versatility—they amplify all sounds rather than just those the wearer wants to hear [4,5]. The perception and translation of speech in a noisy environment are difficult, even When using state-of-the-art devices [6], which necessitates the use of effective and bettermethods to benefit people with hearing difficulties [7,8].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
