Review on Recent Advances in Hearing Aids: A Signal Processing Perspective

Devi, R. Vanitha; VASUNDHARA, .

doi:10.1007/978-981-19-8742-7_19

Cited by 6 publications

(3 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A. RELATED WORK The most notable reviews of AE using DNNs are [20], [4], [6], and [21] to [28]. While [20], [4], and [6] discuss AE applications restricted to speech and music based on a variety of DNNs, the model in [21] is focused only on Deep Reinforcement Learning (DRL) models covering a wide range of applications including Human-Robot Interaction (HRI), music listening and generation, AE, emotions modeling, spoken dialogue systems and automatic speech recognition.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Audio Enhancement Algorithms for Music, Speech, Bioacoustics, Biomedical, Industrial, and Environmental Sounds by Image U-Net

Gul,

Khan

2023

IEEE Access

View full text Add to dashboard Cite

The recent surge in the use of Deep Neural Networks (DNNs) has also made its mark in the field of Audio Enhancement (AE), providing much better quality than the classical methods. Although, there are dedicated audio processing DNNs, yet, many recent models of AE have utilized U-Net: a DNN based on Convolutional Neural Network (CNN), fundamentally developed for image segmentation. It is found that the useful features hidden in the time domain are highlighted when the audio signal is converted to a spectrogram, which can be treated as an image. In this article, we will review the recent work, utilizing U-Nets for different AE applications. Different than other published reviews, this review focuses entirely on AE techniques based on image U-Nets. We will discuss the need for AE, U-Net comparison to other DNNs, the benefits of converting the audio to 2D, input representations that are useful for different AE applications, the architecture of vanilla U-Net and the pre-trained models, variations in vanilla architecture incorporated in different E models, and the state-of-the-art AE algorithms based on U-Net in various applications. Apart from speech and music, this article discusses a wide range of audio signals e.g. environmental, biomedical, bioacoustics, and industrial sounds, not covered collectively in a single article in previously published studies. The article ends with the discussion of colored spectrograms in future AE applications.INDEX TERMS CNNs; image processing deep neural networks; pre-trained networks; spectrogram.;U-Net.

show abstract

Section: Introductionmentioning

confidence: 99%

“…The work in [27] reviews only the source separation application of AE using independent vector analysis. The review in [28] is focused only on machine learning and deep learning models used for AE in hearing aids.…”

Section: Introductionmentioning

confidence: 99%

A Survey of Audio Enhancement Algorithms for Music, Speech, Bioacoustics, Biomedical, Industrial, and Environmental Sounds by Image U-Net

Gul,

Khan

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…As both user safety and user experience are highly important aspects of any new emerging technology, this concern should be addressed in the further development of the instrumented earbuds and the guided exploration training. Possible solutions could be to reduce the passive noise-cancelling effect of the earbuds, to implement technologies currently used in hearing aids (Vanitha Devi & Vasundhara, 2023), or to explore bone conduction headsets (Surendran, Prodanovic, & Stenfelt, 2023). Alternatively, other modalities, such as visual instructions and feedback using augmentedreality glasses could be explored.…”

Section: Discussion and Future Perspectivesmentioning

confidence: 99%

Exploring running-style modulation

Nijs

View full text Add to dashboard Cite

Technological advances in the past decades allow runners to measure a large range of variables by means of wearables and smartphone applications. However, the interpretation of such variables is often lacking. In this thesis, I present a series of foundational studies to implement the concept of guided running-style exploration in the actual training of running. For the development of a successful application based on this concept, three things are required: an appropriate conceptual framework, an accurate measurement system, and effective instructions. The conceptual framework I chose is the dual-axis model, which states that the running style at a certain speed can be described by means of the cadence and duty factor (relative stance time). Thus, cadence, duty factor, and speed are the running variables of interest. For the measurement system, I analysed the test-retest reliability, face validity and concurrent validity of the cadence and stance time derived from instrumented wireless earbuds and found them to be good. This implies that the wireless instrumented earbuds can be used as a valid and reliable measurement system for runners. The core of this thesis is researching effective instructions to modulate cadence and duty factor. For the cadence, acoustic pacing can be a feasible instruction method. Acoustic pacing can be step-based (i.e. a cue for each step) or stride-based (i.e. a cue for every other step), these two forms were compared in terms of effectiveness and their effect on impact forces. Step-based pacing led to more stable auditory-motor coordination than stride-based pacing in both walking and running. There was no effect of the presence of the acoustic pacing on the impact forces. Therefore, acoustic step-based pacing can be used to modulate cadence with the underlying aim of reducing impact forces, without introducing a counterproductive effect. For the duty factor, I tested four verbal instructions: increase or decrease stance time and increase or decrease flight time. Results showed that the participants were able to modulate duty factor without changing cadence. There was no difference between the different instructions, indicating that either type of instruction can be used to modulate the duty factor. When provided separately in overground running, the acoustic pacing and verbal instructions did lead to covarying effects, especially on speed. However, when the acoustic pacing and verbal instructions were combined for running-style modulation according to the dual-axis model, the covarying effects on speed were mitigated. The combined instructions can thus effectively be used for running-style modulation according to the dual-axis model in overground running. In collaboration with the company Dopple, the newly established instructions were implemented in the Running Buddy application for guided running-style exploration. This application is a good first step towards guided running-style exploration, but improvements can still be made, especially in terms of user autonomy. Overall this thesis provides a base for the necessary steps towards guided running-style exploration. With guided running-style exploration, runners can in the future actively vary their running style to reach their personal goals in terms of performance and injury risk.

show abstract

Curvelet based robust improved sine adaptive filter for feedback cancellation in hearing aids

Devi

VASUNDHARA

2023

Applied Acoustics

View full text Add to dashboard Cite

Review on Recent Advances in Hearing Aids: A Signal Processing Perspective

Cited by 6 publications

References 72 publications

A Survey of Audio Enhancement Algorithms for Music, Speech, Bioacoustics, Biomedical, Industrial, and Environmental Sounds by Image U-Net

A Survey of Audio Enhancement Algorithms for Music, Speech, Bioacoustics, Biomedical, Industrial, and Environmental Sounds by Image U-Net

Exploring running-style modulation

Curvelet based robust improved sine adaptive filter for feedback cancellation in hearing aids

Contact Info

Product

Resources

About