[AUDITORY] HearMus Seminar Series (Alinka Greasley )

Subject: [AUDITORY] HearMus Seminar Series From: Alinka Greasley <0000033f2edfad3c-dmarc-request@xxxxxxxx> Date: Thu, 25 Sep 2025 17:25:33 +0000 --_000_AM9PR03MB77112F838DA9AAD93413BDD88B1FAAM9PR03MB7711eurp_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable We are delighted to be joined by Vinzenz Sch=F6nfelder (Mimi Hearing Techno= logies GmbH) and Dr Pauline Mouawad (UCL Ear Institute) for our next instal= ment of the HearMus seminar series which provides a monthly forum for the d= iscussion of a broad array of topics around music and hearing health. All w= elcome! ************************************************************ HearMus Seminar 8 - Wednesday 8th October 14:00-16:00 GMT (15:00-17:00 CEST= ) (Zoom details below) Hands-on experiences with measuring subjective benefit of music processing = strategies for HI listeners Vinzenz Sch=F6nfelder, Mimi Hearing Technologies GmbH Reliably quantifying the benefit of signal processing strategies represents= an essential part of developing products for hearing impaired listeners. H= owever, unlike in the speech processing context, there is no standard metho= d or metric to determine the extent to which the enhancement of music signa= ls improves the individual listening experience. In addition to the specifi= c differences between speech and music stimuli, this is primarily due to di= fferences in the purpose of speech vs. music processing: preference for one= or another music signal is highly subjective and individual, and thus inhe= rently hard to quantify. In this presentation, we will provide an overview over the specific challen= ges and potential solutions in the context of measuring subjective preferen= ce for processing strategies aimed at music signals. We will present and di= scuss study designs and data analysis methods that we have used to quantify= processing benefit and share some study results along with our hands-on ex= perience in the general context of music processing for hearing impaired li= steners. Musical complexity governs a tradeoff between reliability and dimensionalit= y in the neural code Dr Pauline Mouawad, UCL Ear Institute Previous studies have explored neural responses to simple musical sounds, b= ut the neural coding of complex music remains unexplored. We addressed this= gap by analyzing multi-unit activity (MUA) recorded from the inferior coll= iculus (IC) of normal-hearing (NH) and hearing-impaired (HI) gerbils in res= ponse to a range of music types at multiple sound levels. The music types i= ncluded individual stems (vocals, drums, bass, and other) as well as mixtur= es in which the stems were combined. Using coherence analysis, we assessed = how reliably music is encoded in the IC across repeated presentations and t= he degree to which individual stems are distorted when presented in a mixtu= re. To explore neural activity patterns at the network level, we used PCA t= o identify the signal manifold, the subspace where reliable musical informa= tion is embedded. To model neural trans- formations, we developed a deep ne= ural network (DNN) to generate MUA from sound. To assess the impact of hear= ing loss, we compared results for NH and HI at equal sound and sensation le= vels. We identified strong nonlinear interactions between stems, affecting = both the reliability and geometry of neural coding. The reliability of resp= onses and the dimensionality of the signal manifold varied widely across mu= sic types, with reliability decreasing and dimensionality increasing with i= ncreasing musical complexity. The leading modes in the signal manifold were= reliable and shared across all music types, but as musical complexity incr= eased, new neural modes emerged, though these were increasingly unreliable.= Our DNN successfully synthesized MUA from music with high fidelity. After = hearing loss, neural coding was strongly distorted at equal sound level but= these distortions were largely corrected at equal sensation level. Music p= rocessing in the early auditory pathway involves nonlinear interactions tha= t shape the neural representation in complex ways. The signal manifold cont= ains a fixed set of leading modes that are invariant across music types. As= music becomes more complex the manifold is not reconfigured; instead, new,= less reliable modes are added. These new modes reflect a fundamental trade= -off between fidelity and complexity in the neural code. The fact that suit= able amplification restores near-normal neural coding suggests that mild-to= -moderate hearing loss primarily affects audibility rather than the brainst= em's capacity to process music. ************************************************************ Zoom details: Topic: HearMus Seminar 8 (Wed 8th Oct 2-4pm BST 3-5pm CEST) Time: Oct 8, 2025 02:00 PM London Join Zoom Meeting https://universityofleeds.zoom.us/j/82380577236 Meeting ID: 823 8057 7236 ************************************************************ The HearMus seminar series provides a monthly forum for the discussion of a= broad array of topics around music and hearing health. The series aims to = yield a state of the art of research on music perception and hearing impair= ment, hearing aids and music, and individual differences in music perceptio= n and production. Besides presentations from experts in academia and indust= ry, the seminar series seeks to foster lively discussions and exchange of i= deas, with the joint goal of sustaining and enhancing access to music for p= eople with diverse hearing needs. You can access recordings of the previous= seminars here: https://musicandhearingaids.org/hearmus-seminars/ Seminar leaders: Prof. Alinka Greasley, Prof. Kai Siedenburg Prof. Alinka Greasley Professor of Music Psychology Director of Research and Innovation School of Music | University of Leeds | Leeds, LS2 9JT, UK Email: a.e.greasley@xxxxxxxx<mailto:a.e.greasley@xxxxxxxx> | Phone: += 44 113 343 4560 --_000_AM9PR03MB77112F838DA9AAD93413BDD88B1FAAM9PR03MB7711eurp_ Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr= osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" = xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:= //www.w3.org/TR/REC-html40"> <head> <meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Diso-8859-= 1"> <meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)"> <style></style> </head> <body lang=3D"EN-GB" link=3D"#467886" vlink=3D"#96607D" style=3D"word-wrap:= break-word"> <div class=3D"WordSection1"> We are delighted to be joined by Vinzenz Sch=F6nfeld= er (Mimi Hearing Technologies GmbH) and Dr Pauline Mouawad (UCL Ear Institu= te) for our next instalment of the HearMus seminar series which provides a = monthly forum for the discussion of a broad array of topics around music and hearing health. All welcome! <spa= n style=3D"font-size:11.0pt;font-family:"Calibri",sans-serif"> <o:p></o:p> <o:p> </o:p> ****************************************************= ********<o:p></o:p> HearMus Seminar 8 – Wednesday 8th October 1= 4:00-16:00 GMT (15:00-17:00 CEST) (Zoom details below)<o:p></o:p> <o:p> </o:p> Hands-on experiences with measuring subjective be= nefit of music processing strategies for HI listeners<o:p></o:p> Vinzenz Sch=F6nfelder, Mimi Hearing Technologies Gmb= H<o:p></o:p> <o:p> </o:p> Reliably quantifying the benefit of signal processin= g strategies represents an essential part of developing products for hearin= g impaired listeners. However, unlike in the speech processing context, the= re is no standard method or metric to determine the extent to which the enhancement of music signals improves= the individual listening experience. In addition to the specific differenc= es between speech and music stimuli, this is primarily due to differences i= n the purpose of speech vs. music processing: preference for one or another music signal is highly subjectiv= e and individual, and thus inherently hard to quantify.<o:p></o:p> In this presentation, we will provide an overview ov= er the specific challenges and potential solutions in the context of measur= ing subjective preference for processing strategies aimed at music signals.= We will present and discuss study designs and data analysis methods that we have used to quantify processing= benefit and share some study results along with our hands-on experience in= the general context of music processing for hearing impaired listeners.<o:= p></o:p> <o:p> </o:p> Musical complexity governs a tradeoff between rel= iability and dimensionality in the neural code<o:p></o:p> Dr Pauline Mouawad, UCL Ear Institute <o:p></o:p></p= > <o:p> </o:p> Previous studies have explored neural responses to s= imple musical sounds, but the neural coding of complex music remains unexpl= ored. We addressed this gap by analyzing multi-unit activity (MUA) recorded= from the inferior colliculus (IC) of normal-hearing (NH) and hearing-impaired (HI) gerbils in response to a = range of music types at multiple sound levels. The music types included ind= ividual stems (vocals, drums, bass, and other) as well as mixtures in which= the stems were combined. Using coherence analysis, we assessed how reliably music is encoded in the IC ac= ross repeated presentations and the degree to which individual stems are di= storted when presented in a mixture. To explore neural activity patterns at= the network level, we used PCA to identify the signal manifold, the subspace where reliable musical infor= mation is embedded. To model neural trans- formations, we developed a deep = neural network (DNN) to generate MUA from sound. To assess the impact of he= aring loss, we compared results for NH and HI at equal sound and sensation levels. We identified strong no= nlinear interactions between stems, affecting both the reliability and geom= etry of neural coding. The reliability of responses and the dimensionality = of the signal manifold varied widely across music types, with reliability decreasing and dimensionality increas= ing with increasing musical complexity. The leading modes in the signal man= ifold were reliable and shared across all music types, but as musical compl= exity increased, new neural modes emerged, though these were increasingly unreliable. Our DNN successfully s= ynthesized MUA from music with high fidelity. After hearing loss, neural co= ding was strongly distorted at equal sound level but these distortions were= largely corrected at equal sensation level. Music processing in the early auditory pathway involves nonlin= ear interactions that shape the neural representation in complex ways. The = signal manifold contains a fixed set of leading modes that are invariant ac= ross music types. As music becomes more complex the manifold is not reconfigured; instead, new, less reliable mode= s are added. These new modes reflect a fundamental trade-off between fideli= ty and complexity in the neural code. The fact that suitable amplification = restores near-normal neural coding suggests that mild-to-moderate hearing loss primarily affects audibility r= ather than the brainstem’s capacity to process music.<o:p></o:p> <o:p> </o:p> ****************************************************= ********<o:p></o:p> Zoom details:<o:p></o:p> Topic: HearMus Seminar 8 (Wed 8th Oct 2-4pm BST 3-5p= m CEST)<o:p></o:p> Time: Oct 8, 2025 02:00 PM London<o:p></o:p> Join Zoom Meeting<o:p></o:p> <a href=3D"https://universityofleeds.zoom.us/j/82380= 577236">https://universityofleeds.zoom.us/j/82380577236</a><o:p></o:p> Meeting ID: 823 8057 7236<o:p></o:p> <o:p> </o:p> ****************************************************= ********<o:p></o:p> The HearMus seminar series provides a monthly forum = for the discussion of a broad array of topics around music and hearing heal= th. The series aims to yield a state of the art of research on music percep= tion and hearing impairment, hearing aids and music, and individual differences in music perception and product= ion. Besides presentations from experts in academia and industry, the semin= ar series seeks to foster lively discussions and exchange of ideas, with th= e joint goal of sustaining and enhancing access to music for people with diverse hearing needs. You can access reco= rdings of the previous seminars here: <a href=3D"https://musicandhearingaids.org/hearmus-seminars/">https://music= andhearingaids.org/hearmus-seminars/</a><o:p></o:p> <o:p> </o:p> Seminar leaders: Prof. Alinka Greasley, Prof. Kai Si= edenburg<o:p></o:p> <o:p> </o:p> Prof. Alinka Greasley<o:p></o:p> Professor of Music Psychology<o:p></o:p> Director of Research and Innovation<o:p></o:p> School of Music | University of Leeds | Leeds, LS2 9JT, UK= <o:p></o:p> Email: <a href=3D"mailto:a.e.g= reasley@xxxxxxxx">a.e.greasley@xxxxxxxx= </a> | Phone: + 44 113 343 4560</= span><o:p></o:p> <o:p> </o:p> </div> </body> </html> --_000_AM9PR03MB77112F838DA9AAD93413BDD88B1FAAM9PR03MB7711eurp_--

This message came from the mail archive
postings/2025/
maintained by:

DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University