web analytics

Category: Acoustics

Categories
Acoustics Attention Auditory Neuroscience Events Media Neural Filters Publications Speech perception

Mar­tin Orf receives EUHA Award for Out­stand­ing Thesis

Mar­tin Orf is one of three recip­i­ents of this year’s EUHA Award for Out­stand­ing The­sis from the Euro­pean Union of Hear­ing Acousti­cians. His the­sis, titled “Selec­tive Atten­tion in Mul­ti-Talk­er Sit­u­a­tions: Neur­al and Behav­ioral Mech­a­nisms”, offers valu­able insights into the neur­al and behav­iour­al process­es behind selec­tive atten­tion in com­plex lis­ten­ing envi­ron­ments. A key find­ing of his research is that the neur­al rep­re­sen­ta­tion of attend­ed speech becomes stronger when a com­pet­ing, ignored speech stream is being com­pressed in its dynam­ics (a very com­mon yet ill-under­stood sig­nal pro­cess­ing tech­nique in audio pro­duc­tion and also in hear­ing devices). Martin’s dis­cov­ery could con­tribute to the devel­op­ment of future hear­ing aid algo­rithms and in the refine­ment  of exist­ing ones.

Categories
Acoustics Neural Filters Neural Phase Papers Perception Publications Uncategorized

New paper in Devel­op­men­tal Cog­ni­tive Neu­ro­science, Jessen et al.

Our lab (senior author Sarah Tune) teamed up once again with the Baby­lab Lübeck, led by Sarah Jessen: Sarah and Sarah co-wrote a great tuto­r­i­al on how the ver­sa­tile analy­sis frame­work of tem­po­ral response func­tions can be used to analyse brain data obtained in infants. The arti­cle has now been accept­ed for pub­li­ca­tion in the well-reput­ed jour­nal Devel­op­men­tal Cog­ni­tive Neu­ro­science:

 

Categories
Acoustics Familiarity Papers Perception Publications Voice

New Paper in Cog­ni­tion by Lavan, Kre­it­e­wolf et al.

Con­grat­u­la­tions to for­mer Obleser post­doc Jens Kre­it­e­wolf (now at McGill Uni­ver­si­ty) for his new paper in Cog­ni­tion, “Famil­iar­i­ty and task con­text shape the use of acoustic infor­ma­tion in voice iden­ti­ty perception”! 

Togeth­er with our col­leagues from Lon­don, Nadine Lavan and Car­olyn McGet­ti­gan, we took a new approach to test the long­stand­ing the­o­ret­i­cal claim that lis­ten­ers dif­fer in their use of acoustic infor­ma­tion when per­ceiv­ing iden­ti­ty from famil­iar and unfa­mil­iar voic­es. Unlike pre­vi­ous stud­ies that have relat­ed sin­gle acoustic fea­tures to voice iden­ti­ty per­cep­tion, we linked lis­ten­ers’ voice-iden­ti­ty judg­ments to more com­plex acoustic representations—that is, the spec­tral sim­i­lar­i­ty of voice  record­ings (see Fig­ure below).

This new study has a direct link to pop cul­ture (by cap­ti­laz­ing on nat­u­ral­ly-vary­ing voice record­ings tak­en from the famous TV show Break­ing Bad) and chal­lenges tra­di­tion­al pro­pos­als that view famil­iar and unfa­mil­iar voice per­cep­tion as being dis­tinct at all times.

Click here to find out more.