Hearing Systems works with following research fields:
• Auditory signal processing and modeling of auditory perception
• Computational neuroscience of hearing
• Audiology
• Objective measures of auditory function
• Speech perception
• Audio-visual integration
• Auditory scene analysis and communication in complex sound environments
• Hearing instrument signal processing
1) Auditory signal processing and modeling of auditory perception
The research in auditory signal processing and perception is mainly concerned with understanding the relation between basic auditory functions and measures of sound perception such as speech, music or basic sound attributes. Computational models of auditory perception are developed based on the results from listening experiments both in normal-hearing and hearing-impaired listeners. These models are integrated into speech and hearing-aid signal processing applications.
2) Computational neuroscience of hearing
One important aim of computational auditory neuroscience is to model human perception with the help of biological networks. The methods employed to study the neural basis of sound perception are related to nonlinear signal processing, information theory, adaption and learning. Of interest are neurons located along the auditory pathway between the inner ear and the auditory cortex.
3) Audiology
Audiology is concerned with the study of impaired hearing and the possibilities of compensation by means of e.g. a hearing aid. A major topic in this area is to develop new or better diagnostic tools in order to distinguish between different kinds of hearing losses. Traditional descriptions in terms of the audiogram, i.e., pure tone thresholds, have been shown to be far from satisfying. Better diagnostic tools are essential for the fitting of advanced signal processing hearing aids, thus allowing the hearing-impaired user to obtain the benefit from the device. Identifying and quantifying factors describing hearing loss are therefore one of the main goals of the research into audiology.
4) Objective measures of auditory function
The field of objective measures of auditory function comprises studies of auditory evoked potentials and otoacoustic emissions. Auditory evoked potentials represent electrical fields recorded from the surface of the head in response to sound. Otoacoustic emissions are low-level sounds generated by the inner ear, the cochlea, in response to sound or even in the absence of external sound (spontaneous activity), and can be recorded in the ear canal. Otoacoustic emissions and evoked potentials can provide insights into the physiological state of the ear and the brain. Research projects range from basic scientific interpretations of the objective measures themselves to applied technical work on improving signal quality and robustness under noisy conditions.
5) Speech perception
The research into speech perception aims at describing the mechanisms underlying speech intelligibility, i.e. how human listeners decode and integrate the information carried by the speech signal. Moreover, models of speech intelligibility and models quantifying speech perception under challenging listening conditions are under development. Like other areas in the group, the research is interdisciplinary and relies on psychoacoustics, auditory signal processing and phonetics.
6) Audio-visual integration
The research into the field of audio-visual integration is concerned with the integration process of visual cues (e.g. lip reading) and auditory cues performed in the brain. It includes synchronized audio-visual speech generation (“talking heads”) and investigations into neural synchronization in the brain. The central goal is to design a computer-based ‘language laboratory’ for lipreading that can be used by people with a sudden profound hearing loss to quickly re-gain communicational skills. Visual speech information is also useful for human-machine communication in noisy environments.
7) Auditory scene analysis and communication in complex sound environments
Computational auditory scene analysis (CASA) is a research field that recently has had its focus on applications, such as automatic speech recognition systems. The aim of our activities is to use CASA to achieve a better understanding of the principles of human hearing underlying the perception of complex acoustic scenes.
8) Hearing instrument signal processing
Modern hearing instruments apply a large number of signal processing algorithms, such as noise-reduction and feedback cancellation. Different strategies to compensate for a patient’s hearing impairment are applied. We aim at investigating and contributing to the development of new algorithms and compensation strategies, based on the most recent findings about the source and consequence of hearing impairment.