Hearing Seminars
CCRMA hosts a weekly Hearing seminar. All areas related to perception are discussed, but the group emphasizes topics that will help us understand how the auditory system works. Speakers are drawn from the group and visitors to the Stanford area. Most attendees are graduate students, faculty, or local researchers interested in psychology, music, engineering, neurophysiology, and linguistics. Meetings are usually from 10:30AM to 12:20 (or so, depending on questions) on Friday mornings in the CCRMA Seminar Room.
The current schedule is announced via a mailing list. To be added to the mailing list, send email to hearing-seminar-request@ccrma.stanford.edu. If you have any questions, please contact Malcolm Slaney at hearing-seminar-admin@ccrma.stanford.edu.
Recent Hearing Seminars
Listening effort depends on acoustic cues the brain uses for sound segregation and selective attention
Date:Fri, 11/11/2016 - 10:30am - 12:00pmLocation:CCRMA Seminar RoomEvent Type:Hearing SeminarThere are many cues in our acoustic environment that we integrate to help us understand the world around us. One of the most important sources of information are spatial cues. Sources that we wish to distinguish are often spatially separated, and sounds that we want to integrate almost always come from a single direction. But what happens when the cues are conflicting? This might happen with bad stereo reproduction, or with bad hearing aids.
FREEOpen to the PublicEstimating pitch with time-domain DNNs
Date:Fri, 11/04/2016 - 10:30am - 12:00pmLocation:CCRMA Seminar RoomEvent Type:Hearing SeminarCCRMA Student Prateek Verma will be discussing his work on estimating pitch using deep neural networks (DNNs) in the time domain.
Without a doubt, the most important recent advancement in machine learning (and AI) has been the success of deep neural networks. They are key to modern speech recognition and many other areas.FREEOpen to the PublicUsing Listening Effort to Measure Audio Quality
Date:Fri, 10/28/2016 - 10:30am - 12:00pmLocation:CCRMEvent Type:Hearing SeminarThere are many ways to measure audio quality. In the past we have used engineering measures such as the mean-squared error, or simply asked people. But these measures are either cheap engineering hacks without much basis in perception, or expensive to measure and not very accurate.
Instead, our friends in Berkeley have redefined the problem in terms of listening effort. How *hard* is it to understand the audio, especially as we go about our business? Do you enjoy listening to the audio? Do you have to concentrate on the audio to understand it? How much work does it take to understand the audio signal in noisy and multi-speaker situations?FREEOpen to the PublicEd Large on Beat Tracking Like the Brain
Date:Fri, 10/21/2016 - 10:30am - 12:00pmLocation:CCRMA Seminar RoomEvent Type:Hearing SeminarNeural population rhythms (oscillations) are cyclical fluctuations of baseline neuronal activity that can be observed in neocortical and thalamic regions of the brain. Auditory rhythmic stimulation can entrain auditory cortical rhythms and broadly distributed motor rhythms. Auditory-motor entrainment is thought to underlie pulse and meter perception in music and enable coordination of rhythmic movements with musical rhythms. Auditory-motor coupling is thought to be reciprocal and plastic. I will describe a recent dynamical model of cortical synchronization to complex rhythms using gradient frequency neural networks. The model successfully predicts neural synchronization and synchronized tapping to “missing pulse” rhythms.
FREEOpen to the PublicIntroduction and Review of Auditory Saliency
Date:Fri, 10/07/2016 - 10:30am - 12:00pmLocation:CCRMA Seminar RoomEvent Type:Hearing SeminarPOW!!!
That was clearly salient. But that sound was easy. What, in general, makes a sound salient? How do we model and predict saliency? Saliency is just the first part of auditory scene analysis. A sound must be salient, in some fashion, before we can separate it out into its own sound object.FREEOpen to the PublicSlim Essid on Analysing Audio and EEG data with Nonnegative Matrix Factorisation
Date:Fri, 09/30/2016 - 10:30am - 12:00pmLocation:CCRMA Seminar RoomEvent Type:Hearing SeminarWe are starting to think about audio and EEG signals as combinations of sources. This has been true of both images and audio for a while, and an important breakthrough in these types of analysis is the realization that the sources must be strictly positive. Algorithms such as PCA and the FFT divvy up the energy in a signal using both positive and negative components. Non-negative matrix factorization (NMF) creates an alternative explanation of the data where all the factors are strictly positive, since objects can only contribute positive energy to the overall signal.
Slim Essid will be at CCRMA to talk about using NMF to analyze audio signals for auditory scene analysis, and to decompose EEG signals into their independent sources. Both important tasks.FREEOpen to the PublicJérémie Voix on In-ear Wearable Technologies and Musical Applications
Date:Tue, 02/23/2016 - 5:00pm - 6:15pmLocation:CCRMA Seminar RoomEvent Type:Hearing SeminarWearables are everywhere. But not in the ears (yet). The CRITIAS research team has been actively developing various in-ear technologies designed to complement the human ear, from “intelligent” protection for professional musicians, to advanced inter-individual communication systems, to otoacoustic emission (OAE) monitoring devices or in-ear EEG Brain Computer Interface (BCI). More fundamental research has also been conducted, particularly on the micro-harvesting of electrical power from inside the earcanal to power future auditory wearables. One recent development, the Auditory Research Platform, will be presented to the CCRMA community who may take benefit from its versatility and portability for immersive in-ear audio processing applications.
FREEOpen to the PublicDaniel Strauss on Objective Estimation of Listening Effort with EEG
Date:Wed, 01/13/2016 - 10:30am - 12:00pmLocation:CCRMA Seminar RoomEvent Type:Hearing Seminar
We have analyzed the attentional dimension of listening effort using electroencephalographic methods and quantitative neurofunctional modeling in recent years. In particular, we have developed a neurophysical corticothalamic feedback model for listening effort correlates in event-related potentials (ERP) which allowed us to compare simulated and measured data. However, the resulting ERP paradigms are restricted to a certain class of stimulation protocols, limiting their applications, e.g., EEG-aided hearing instrument fitting.
FREEOpen to the PublicHideki Kawahara on STRAIGHT (high resolution speech modifications)
Date:Fri, 11/20/2015 - 10:30am - 12:00pmLocation:CCRMA Seminar RoomEvent Type:Hearing Seminar
Prof. Kawahara has spent many years building STRAIGHT, an ultra-high resolution approach to analyzing and modifying speech signals, that is the basis of many speech manipulation experiments and products. He will be at CCRMA to discuss STRAIGHT, its history, its approach, and current status.
FREEOpen to the PublicProf. Takako Fujioka on Neural correlates for prediction of musical structures and learning-related functional plasticity
Date:Fri, 11/13/2015 - 10:30am - 12:00pmLocation:CCRMA Seminar RoomEvent Type:Hearing Seminar
What can EEG (and MEG) tell us about how we perceive polyphony, syntax, and longer-scale relationships? How does our perception of time develop as we listen to a musical piece? I think this work is important because we know precious little about how we music perception works, at the neural level. EEG and MEG give us tools to help us understand how we represent music internally, and how our perception develops over time. I’m very happy to announce that Prof. Takako Fujioka will be at the Hearing Seminar to update us with her latest work.
Who: Prof. Takako Fujioka (Stanford CCRMA)FREEOpen to the Public
