Eric Durant, Ph.D., MBA, P.E.
Professor, Program Director
- Milwaukee WI UNITED STATES
- Walter Schroeder Library L339
- Electrical Engineering and Computer Science
Dr. Eric Durant's research focuses on real-time audio processing with a focus on hearing aids and artificial intelligence/deep learning.
Multimedia
Education, Licensure and Certification
Professional Engineer
WI License 45011-6
Executive M.B.A.
Business
University of Wisconsin-Milwaukee
2011
Ph.D.
Electrical Engineering
University of Michigan
2002
M.S.E.
Electrical Engineering
University of Michigan
1999
B.S.
Computer Engineering
Milwaukee School of Engineering
1998
B.S.
Electrical Engineering
Milwaukee School of Engineering
1998
Biography
Areas of Expertise
Accomplishments
Order of the Engineer, MSOE inductee #500
2019
MSOE Alumni Achievement Award
2017
Oscar Werwath Distinguished Teacher Award, MSOE
2016
MSOE Inaugural Student Athletic Advisory Committee (SAAC) Faculty Recognition Award
Awarded for contributions in MSOE sports photography, May, 2015
STEM Forward Young Engineer of the Year
2013
EECS Department Fellowship
University of Michigan, 1998-2002
Affiliations
- American Society for Engineering Education (ASEE) : Member
- Institute of Electrical and Electronics Engineers (IEEE) : Senior Member
- ABET PEV : Computer and Electrical Engineering Program Evaluator
Social
Media Appearances
Faculty and staff honored at MSOE
MSOE
2016-05-09
Dr. Eric Durant ’98, professor and computer engineering program director, received the Oscar Werwath Distinguished Teacher Award. The award was established by the university in 1967 to recognize excellence in teaching. All nominees for this award must have a minimum of seven years of full-time service to MSOE. Students choose the award winner through two rounds of voting. On their ballots, students described Durant as a passionate teacher who clearly wishes for students’ success; extremely dedicated to being the best educator and mentor he can be; incredibly nice and humble; helpful to students, even those he hasn’t met before; by far the most caring professor I have ever had; and he always makes times to answer questions outside of class.
Patents
Intelligent spoken command response systems
US Patent 11412333
2022
In an audio signal, one or more processing circuits recognize spoken content in a user's own speech signal using speech recognition and natural language understanding. The spoken content describes a listening difficulty of the user. The one or more processing circuits generate, based on the spoken content, one or more actions for hearing devices and feedback for the user. The one or more actions attempt to resolve the listening difficulty. Additionally, the one or more processing circuits convert the user feedback to verbal feedback using speech synthesis and transmit the one or more actions and the verbal feedback to the hearing devices via a body-worn device. The hearing devices are configured to perform the one or more actions and play back the verbal feedback to the user.
Method and apparatus for localization of streaming sources in hearing assistance system
US9930456B2
2018
A hearing assistance system streams audio signals from one or more streaming sources to a hearing aid set and enhances the audio signals such that the output sounds transmitted to the hearing aid wearer include a spatialization effect allowing for localization of each of the one more streaming sources. The system determines the position of the hearing aid set relative to each streaming source in real time and introduces the spatialization effect for that streaming source dynamically based on the determined position, such that the hearing aid wearer can experience a natural feeing of the acoustic environment.
Method and apparatus for localization of streaming sources in hearing assistance system
US9584933B2
2017
A hearing assistance system streams audio signals from one or more streaming sources to a hearing aid set and enhances the audio signals such that the output sounds transmitted to the hearing aid wearer include a spatialization effect allowing for localization of each of the one more streaming sources. The system determines the position of the hearing aid set relative to each streaming source in real time and introduces the spatialization effect for that streaming source dynamically based on the determined position, such that the hearing aid wearer can experience a natural feeing of the acoustic environment.
Method and apparatus for localization of streaming sources in hearing assistance system
US9124983B2
2015
A hearing assistance system streams audio signals from one or more streaming sources to a hearing aid set and enhances the audio signals such that the output sounds transmitted to the hearing aid wearer include a spatialization effect allowing for localization of each of the one more streaming sources. The system determines the position of the hearing aid set relative to each streaming source in real time and introduces the spatialization effect for that streaming source dynamically based on the determined position, such that the hearing aid wearer can experience a natural feeing of the acoustic environment.
Hearing aids and methods and apparatus for audio fitting thereof
US9049529
2015
A field ready, unsupervised-use ready, method and apparatus for audio fitting a hearing aid is described in a hand held configuration having paired comparisons (hearing selections) stored in and derivable from a memory therein. The paired comparisons are presented one at a time to a user and a preferred selection for each paired comparison is made by a select indicator after the user toggles back and forth between the selections for as many times necessary in determining their preferences. A genetic algorithm converges all the preferences upon a single solution. Crossover and mutation genetic algorithm operators operate on a linear range of indexes representative of parametric values of the pairs. A fully integrated hearing aid having all the above described features incorporated therein is also presented.
Genetic algorithms with robust rank estimation for hearing assistance devices
US8,359,283
2013
Disclosed herein, among other things, is a method for fitting a hearing assistance device for a user using a genetic algorithm. Stimulus pairs are presented to the user using a computer, the stimulus pairs adapted to provide contrasting options for selection from a set of stimuli stored in the computer. Inputs are received from the user entered into the computer, including preference judgments of the user. A score is calculated for each stimulus of the pair using the computer to execute a rank agreement function to maximize agreement between scores and the preference judgments. A set of genes is selected based on the scores, where the set of genes correspond to hearing assistance device parameters. The set of genes is operated on with a genetic algorithm using the assigned scores to obtain a child set of genes. The child set is used to provide parameter values during fitting.
Efficient convex optimization for real-time robust beamforming with microphone arrays
US8824711
2011
Disclosed herein, among other things, are methods and apparatus for improving speech intelligibility for speech-in-noise in audio processing and hearing assistance devices. The present subject matter includes a method for improving speech intelligibility for speech-in-noise for audio processing and hearing assistance devices. The method includes receiving an audio signal using a microphone array and processing the received signal to improve speech intelligibility in noise. A barrier-type beamforming process is used to improve signal-to-noise ratio at the output of the microphone array. The beamforming process includes convex optimization using a logarithmic barrier function, according to various embodiments.
Hearing aids and methods and apparatus for audio fitting thereof
US7,650,004
2010
A field ready, unsupervised-use ready, method and apparatus for audio fitting a hearing aid is described in a hand held configuration having paired comparisons (hearing selections) stored in and derivable from a memory therein. The paired comparisons are presented one at a time to a user and a preferred selection for each paired comparison is made by a select indicator after the user toggles back and forth between the selections for as many times necessary in determining their preferences. A genetic algorithm converges all the preferences upon a single solution. Crossover and mutation genetic algorithm operators operate on a linear range of indexes representative of parametric values of the pairs. A fully integrated hearing aid having all the above described features incorporated therein is also presented.
Research Grants
Computer and Software Engineering Curricula Development Workshops
NSF Grant 1338752
2013
Co-PI with Mark Ardis of Stevens Institute of Technology
Selected Publications
CE2016: Updated curricular guidelines for computer engineering
IEEE Frontiers in Education Conference (FIE)Nelson, V., Durant, E., Impagliazzo, J., Hughes, J.L.
2017
The report, Curriculum Guidelines for Undergraduate Degree Programs in Computer Engineering (CE2016), developed by the Association for Computing Machinery and the IEEE Computer Society, was released in December of 2016. This is one volume of a series of reports covering curricula for a variety of computing fields; it is a significant update of the previous version, CE2004. This paper discusses significant aspects of CE2016, with a focus on how the report might be used in reviewing, updating, and creating computer engineering programs.
Perceptually motivated ANC for hearing-impaired listeners
IEEEEric Durant, Jinjun Xiao, Buye Xu, Martin McKinney, Tao Zhang
2013
The goal of noise control in hearing aids is to improve listening perception. In this paper we propose modifying a perceptually motivated active noise control (ANC) algorithm by incorporating a perceptual model into the cost function, resulting in a dynamic residual noise spectrum shaping technique based on the time-varying residual noise. The perceptual criterion to be minimized could be sharpness, discordance, annoyance, etc. As an illustrative example, we use loudness perceived by a hearing-impaired listener as the cost function. Specifically, we design the spectrum shaping filter using the listener's hearing loss and the dynamic residual noise spectrum. Simulations show significant improvements of 3-4 sones over energy reduction (ER) for severe high-frequency losses for some common noises that would be 6-12 without processing. However, average loudness across a wide range of noises is only slightly better than with ER, with greater improvements realized with increasing hearing loss. We analyze one way in which the algorithm fails and trace it to over-reliance on the common psychoacoustic modelling simplification that auditory channels are independent to a first approximation. This suggests future work that may improve performance.
Efficient convex optimization for real-time robust beamforming with microphone arrays
IEEEEric Durant, Ivo Merks, Bill Woods, Jinjun Xiao, Tao Zhang, Zhi-Quan Luo
2011
This paper presents an efficient implementation of a robust adaptive beamforming algorithm based on convex optimization for applications in the processing-constrained environment of a digital hearing aid. Several modifications of the standard interior point barrier method are introduced for use where the array data covariance matrix is changing rapidly relative to the algorithm's convergence rate. These efficiency improvements significantly simplify the computation without affecting the algorithm's fast convergence, and are useful for real-time adaptive beamforming regardless of the rate of array correlation change. Simulation results show that this implementation is numerically stable and succeeds where many minimum-variance distortionless response (MVDR) solutions fail.
Rubrics for assessing oral communication in the capstone design experience: Development, application, analysis and refinement
The International Journal of Engineering EducationHenry Welch, Deepti Suri, Eric Durant
2009
The importance of good communication skills is becoming increasingly relevant to engineers in today's globally competitive environment. The Accreditation Board for Engineering and Technology (ABET), recognizing this phenomenon, introduced six professional skills along with the various hard skills in their new accreditation criteria EC2000 for all engineering programs. At the Milwaukee School of Engineering (MSOE), rubrics were developed to aid in assessing the oral presentations made during the capstone senior design sequence. These rubrics have been applied by various senior design professors each quarter to assess all the mid-quarter presentations. The analysis (using the Spearman Rank Correlation Test and a Rater Disagreement Metric) of data collected over four quarters indicates that by repeatedly applying, analyzing and refining a rubric, it is possible to minimize the often subjective means of evaluating communication skills and move towards more objective evaluations. Over the past three years, multiple evaluators have shown strong agreement in the quality of student presentations. However, they have not yet arrived at a complete consensus indicating that we as yet do not have a completely reliable and objective tool and more work needs to be done in this area.
Efficient perceptual tuning of hearing aids with genetic algorithms
IEEEEric A Durant, Gregory H Wakefield, Dianne J Van Tasell, Martin E Rickert
2004
We describe a system for integrating a genetic algorithm (GA) with perceptual feedback to perform an efficient search in a perceptual space. The main system components are an efficient method for estimating perceptual rank order and genetic operators that take advantage of the types of parameters found in certain classes of audio processing systems. Preference judgments are used, resulting in a lightweight user interface. The application to subjectively fitting a portable hearing aid based solely on binary feedback is discussed. An experiment was conducted using eight normal and eight hearing impaired subjects. Three parameters were varied to control cancellation of acoustic feedback. The GA worked well for fitting this system, as indicated by both objective and subjective measures. In addition, users had greatly differing preferences for feedback cancellation parameters and these preferences did not change much when subjects were retested.
Hearing aid fitting with genetic algorithms Authors Eric Alan Durant
semanticscholar.orgEric Alan Durant
2002
Fitting modern DSP-based hearing aids, which have numerous parameters, is a challenge for even the experienced audiologist. Simple but proven fitting rationales are used, limiting the prescribed setting to a subset of the hearing aid’s capabilities [7]. Given the success of the genetic algorithm (GA) in tuning simpler systems with subjective feedback [55, 58, 66], we propose it as an efficient and robust method for fitting hearing aids. GAs borrow biological concepts such as natural selection and mutation and apply them to non-biological search and optimization problems. Many previous studies of genetic algorithms for perceptual tuning [54, 59] and other adaptive hearing aid fitting methods1 [46] were conducted in tightly controlled laboratory environments, limiting their applicability to real-world use. In contrast, our approach uses a simple paired comparison input method, suitable for both clinical and unsupervised field use. This input method provides an easy way to integrate patient input into the fitting procedure, which is recognized as a key to successful fitting by the hearing aid research community [10].
We first investigate whether a GA can guide a hearing-impaired subject towards perceptually good feedback canceller parameters in everyday conditions. Feedback occurs when hearing aid gain is increased so far that the portion of the signal at the microphone due to the hearing aid output is greater than the portion from the environment. Since the output is fed back and amplified further, the hearing aid quickly reaches its maximum output level. This results in a loud squealing sound.
Efficient model fitting using a genetic algorithm: pole-zero approximations of HRTFs
IEEEEric A Durant, Gregory H Wakefield
2002
We demonstrate that a genetic algorithm (GA) can efficiently generate accurate low-order pole-zero approximations of head-related transfer functions (HRTFs) from measured impulse responses by minimizing a logarithmic error criterion. This approach is much simpler and comparable or superior in efficiency to competing search algorithms. We build on previous work in low-order HRTF approximation. By applying the GA, we converge to solutions of equal quality in about 30 s compared to over 20 min. This work develops a basic steady-state GA using a pole-zero filter design problem as an illustrative example. We propose a domain-appropriate error measure. We then apply the algorithm to designing filters to approximate measured HRTFs. Detailed performance measurements are presented. In the appendix, we propose a widely applicable population variation metric. A lower bound is developed for this metric and is used to detect convergence.
