2008 Seminars

January 2008

Seminar with David Perry January 4th

Topic related to Cochlear Implant stimulation and central auditory system changes

David Perry
Department of Otolaryngology, University of Melbourne
and
The Bionic Ear Institute
Melbourne, Australia

February 2008

Seminar with Cameron Morland February 29th

"What it is like to be a Bat: A Sonar system for Humans"

Cameron Morland, Ph.D Candidate
Dept. of Cognitive and Neural Systems
Boston University

March 2008

Seminar with Philip S. Lobel March 7th

"Fish Sex: their sounds of ecstasy fall on our deaf ears"

Philip S. Lobel
Professor, Boston University Marine Program
Department of Biology
Boston University

Seminar with Oded Ghitza March 14th

"On the possible role of brain rhythms in speech perception"

Oded Ghitza, Ph.D.
Center for Biodynamics
Department of Biomedical Engineering
Hearing Research Center
Boston University

Seminar with Roozbeh Ghaffari March 21st

"Longitudinally propagating traveling waves of the mammalian tectorial membrane"

Roozbeh Ghaffari Ph.D Candidate
Research Laboratory of Electronics
Department of Electrical Engineering & Computer Science
Mass. Inst. of Tech.
Cambridge, MA

April 2008

Seminar with Elliot Saltzman April 4th

"A task-dynamic toolkit for modeling the effects of prosodic structure on articulation"

Prof. Elliot Saltzman
Boston University and Haskins Laboratories

Seminar with Erick Gallun April 11th

"Amplitude modulation sensitivity as a mechanism for speech perception"

Erick Gallun
Research Investigator
National Center for Rehabilitative Auditory Research
Portland VA Medical Center
Portland, Oregon

Seminar with Christopher Shera April 18th

"Coherent reflection and its discontents"

Christopher Shera
Eaton-Peabody Laboratory
Mass. Eye and Ear Infirmary
Boston, MA

Seminar with Jyrki Ahveninen April 25th

"Spatiotemporal brain imaging of short-term plasticity of human auditory cortex"

Dr. Jyrki Ahveninen
Department of Radiology, MGH
Harvard Medical School

May 2008

Seminar with David Poeppel May 1st

"Temporal primitives in auditory cognition and speech perception"

ABSTRACT:
Generating usable internal representations of speech input, or an auditory scene more generally, requires, among other operations, fractionating the signals into temporal units/chunks of the appropriate granularity. Adopting (and adapting) Marr's (1982) approach to vision, a perspective is outlined that formulates linking hypotheses between specific neurobiological mechanisms (for example cortical oscillations and phase-locking) and the representations that underlie auditory cognition (for example syllables). Focusing on the implementational and algorithmic levels of description, I argue that the perception of sound patterns requires a multi-time resolution analysis. In particular, recent experimental data from psychophysics, MEG (Luo & Poeppel, 2007), and concurrent EEG/fMRI (Giraud et al., 2007) suggest that there exist two privileged time scales that form the basis for constructing elementary auditory percepts. These "temporal primitives" permit the construction of the internal representations that mediate the analysis of speech and other acoustic signals.

Prof. David Poeppel
Department of Linguistics & Department of Biology
Neuroscience and Cognitive Science Program
University of Maryland, College Park

Seminar with Carol Espy Wilson May 9th

"Speech Enhancement based on the Modified Phase-Opponency Model"

ABSTRACT

In this talk I will discuss a speech enhancement algorithm we have developed that is based on the auditory PO model proposed for detection of tones in noise. The PO model includes a physiologically realistic mechanism for processing the information in neural discharge times and exploits the frequency-dependent phase properties of the tuned filters in the auditory periphery by using a cross-auditory-never-fiber coincidence detection for extracting temporal cues. An important feature of the PO model is that it does not need to estimate the noise characteristics, nor does it assume that the noise satisfies any statistical model. We modified the PO model (MPO) so that its basic functionality was maintained, but the properties of the model can be analyzed and modified independently. In addition, we improved on its performance by coupling the PO model with our Aperiodicity, Periodicity, Pitch (APP) detector. Presently, we are investigating additional processing to further improve the performance of the MPO-APP at low signal-to-noise ratios (6 dB and lower).

Carol Espy Wilson, Ph.D.
Professor, Department of Electrical & Computer Engineering
Institute for Systems Research
University of Maryland

Seminar with John Agapiou May 16th

"Low-frequency envelope sensitivity produces asymmetric binaural tuning curves"

Dr. John Agapiou
Rockefeller University

Seminar with Yuxuan Zhang May 23rd

"Binaural hearing: naive performance and learning"

Yuxuan Zhang
Northwestern University

Seminar with Claus-Peter Richter May 30th

"Activation of neurons with optical radiation - a paradigm shift in neural stimulation?"

ABSTRACT

Neural prosthetic devices are artificial extensions to the body that restore or supplement nervous system function that was lost during disease or injury. The devices stimulate remaining neural tissue providing some input to the nervous system. Hereby, the challenge for neural prostheses is to stimulate remaining neurons selectively. However, the electrical current spread does not allow easily stimulating small neuron populations. In neural prostheses developments, particular success has been realized in the cochlear prostheses development. The devices bypass damaged hair cells in the auditory system by direct electrical stimulation of the auditory nerve. Stimulating discrete spiral ganglion cell populations in cochlear implant users' ears is similar to the encoding of small acoustic frequency bands in a normal-hearing person's ear. In contemporary cochlear implants, however, the injected electric current is spread widely along scala tympani and across turns. Consequently, stimulation of spatially discrete spiral ganglion cell populations is difficult. One goal of implant device development is to design cochlear implants that stimulate smaller populations of spiral ganglion cells. Extreme spatially selective stimulation is possible using optical radiation from a pulsed infrared laser.
The presentation will demonstrate that stimulation of auditory neurons is possible with optical radiation, is spatially selective, and is possible at high repetition rates. The underlying mechanism for stimulating neurons with optical radiation will be discussed.

Prof. Claus-Peter Richter, M.D., Ph.D.
Director of Resident Research
Northwestern University Feinberg School of Medicine

June 2008

Seminar with Michael Mandel June 13th

"Model-based EM Source Separation and Localization in Reverberant Mixtures"

Michael Mandel
Ph.D. Candidate
LabROSA (Laboratory for the Recognition and Organization of Speech and Audio)
Dept. of Electrical Engineering
Columbia University

Seminar with Ram Krips June 20th

"Stochastic Properties of Neural Coincidence Detector cells"

ABSTRACT

Neural information is characterized by sets of spiking events that travel within the brain through neuron junctions that receive, transmit and process streams of spikes. Coincidence detection (CD) is one of the common ways to describe the functionality of a single neural cell.
An analytical derivation of the output stochastic behaviour of a coincidence detector with both excitatory and inhibitory inputs will be presented. The stochastic behaviour of the inputs is described as a non-homogeneous Poisson process (NHPP). The derivation, which is based on an efficient breakdown of the cell into basic functional elements, results in an output process whose behaviour can be approximated as a NHPP as long as the coincidence interval is much smaller than the refractory period of the cell's inputs. Intuitively, the approximation is valid as long as the processing rate is much faster than the incoming information rate. For example an excitatory excitatory (EE) nucleus that receives 2 inputs each behaves as a NHPP with instantaneous rates (IRs) LAMBDA-1 and LAMBDA-2, and a coincidence window (DELTA), then EE output is also NHPP and its IR can be expressed simply (XXX). Similarly EI cell output yields YYYYY, where LAMBDA-E and LAMBDA-I are the excitatory and inhibitory input IRs.
This derivation enables predictions of numerous properties, for example: (1) Decrease in spiking rate at higher brain levels. (2) SNR improvement. (3) Prediction of binaural localization Performance: JND for ITD, ILD and MAA.

Ram Krips, Ph.D. Candidate
School of Electrical Engineering
Tel Aviv University
Tel Aviv, Israel

NOTE: It is my fault that the expressions XXXX and YYYY are missing... Steve Colburn

Seminar with G. Christopher Stecker June 23rd

"Combining binaural cues across time and type"

G. Christopher Stecker, Ph.D.
Assistant Professor
Speech and Hearing Sciences
University of Washington, Seattle

September 2008

Seminar with Bill Hartmann September 9th

Michigan State Physics and Astronomy
[Visiting faculty member, Boston University fall 2008]

"Detecting Interaural Incoherence and the Masking Level Difference"

ABSTRACT:         Human listeners can detect a very small amount of interaural incoherence added to an otherwise completely coherent noise. Experiments with many different narrow-band noises indicate that this detection is based on sensitivity to fluctuations in the interaural phase and in the interaural level because there is good correlation between detection scores and the sizes of these fluctuations. Good correlation like this even occurs in an experimental context where the interaural cross-correlation function is the same for all the different noises. With increasing bandwidth, the fluctuations in different noises become similar and detection scores become similar as well. In experiments where the interaural incoherence so small that it is hard to detect, listeners often make decisions based on the size of fluctuations in the level of the noise because they confuse these level fluctuations with interaural fluctuations. In experiments where the duration of the stimulus is so short that there is no time for fluctuations (bandwidth X duration < 0.5) listeners may make decisions based on the displacement of the noise image from the midline.

Further experiments with narrow-band noises were done to try to discover the rule for combination of interaural phase fluctuations and interaural level fluctuations. The best model for combination was one in which RMS phase fluctuations and RMS level fluctuations were independently computed and then added together. It is conjectured that the masking level difference (MLD), specifically the detection of a heterophasic signal in a homophasic noise, is mediated by the same fluctuation sensitivity. A review of the experimental literature over the sixty-year history of the MLD provides evidence in support of this conjecture.

Seminar with Morten Jepsen September 12th

Centre for Applied Hearing Research
Technical University of Denmark

"Modeling auditory perception in normal-hearing and hearing-impaired listeners"

ABSTRACT:         Advances in auditory modeling have increased the effort on exploiting the knowledge gained from modeling as well as using the output of auditory processing models in applications. Here, a computational auditory signal-processing and perception (CASP) model is described (Jepsen et al., 2008). It is based on the perception model of Dau et al. (1997), and includes nonlinear cochlear processing and other modifications. The inclusion of the compressive cochlear stage allows investigating the perceptual consequences of reduced or lost compression. Results are presented on how to derive model parameters based on psychoacoustic measurements in individual hearing-impaired listeners. It will be discussed how such a model of individual hearing loss may be used in future applications such as hearing-aid development.

Seminar with Oded Ghitza, Ph.D. September 19th

Center for BioDynamics and Hearing Research Center, Boston University
Sensimetrics Corporation

"Consonant Discrimination of Degraded Speech using an Efferent-inspired Closed-loop Cochlear Model"

ABSTRACT:         We present a model of auditory speech processing capable of predicting consonant confusions by normal hearing listeners, based on a phenomenological model of the Medial Olivocochlear efferent pathway. We then use this model to predict human error patterns of initial consonants in consonant-vowel-consonant words. In the process we demonstrate its potential for speech identification in noise. Our results produced performance that was robust to varying levels of additive noise and which was similar to human performance in discrimination of synthetically spoken consonants.
[This talk is based on the Ph.D dissertation of David Messing at MIT (graduated, August 2007).]

Seminar with Douglas Cotanche, Ph.D. September 26th

Hearing Research Center and Department of Otolaryngology
Boston University

"Hair Cell Regeneration and Stem Cell Transplantation in the Vertebrate Inner Ear: Their Potential as Therapeutic Treatments for Deafness"

November 2008

Seminar with Prof. Karen Helfer, Ph.D. November 7th

University of Massachusetts, Amherst

December 2008

Seminar with Rapeechai Navawongse, Ph.D. candidate December 5th

Department of Biomedical Engineering
Boston University

DISSERTATION DEFENSE:         Dissertation Supervisor: Prof. Herbert F. Voigt
Dissertation Defense: "Extracellular Single Neuron Recording in Dorsal Cochlear Nucleus of the Awake Gerbil"

Seminar with Prof. Jens Blauert, Ph.D. December 12th

Ruhr University, Bochum,Germany

"Room-Related Presentation of Auditory Scenes via Loudspeakers"
ABSTRACT:         In this lecture, currently available methods for room-related loudspeaker reproduction are presented, critically discussed and compared. A focus is put on their application in Virtual-Reality systems. The following methods are covered:
Stereophony, as based on
- Amplitude Differences (e.g., Blumlein)
- Additional Time Differences (e.g., ORTF)
Surround Sound, as based on - Amplitude Panning (e.g., Vector Based, First-Order Ambisonics)
- Holophony (e.g., Higher-Order Ambisonics, Wave-Field Synthesis)
- Cue-Selection Procedures (e.g., DirAC)