We are delighted to announce the CfP for the 18th International Workshop on Acoustic Signal Enhancement (IWAENC) 2024, which will be hosted in Aalborg, Denmark from Sep 9 to Sep 12. We invite you to submit your research papers to IWAENC 2024! Read more at iwaenc2024.org.
Our newest work on how to make sound zones robust to the natural pertupations of IRs/ATFs is now available. In the paper, we propose to train models off-line of the ATFs and the ways in which they vary which are then used in the optimal filters for
ieeexplore.ieee.org/document…
180
Mads Græsbøll Christensen retweeted
Tak til @di_digital_ for et indsigtsfuldt indspark i debatten. På AAU arbejder vi hårdt på øge antallet af IT-specialister i 🇩🇰. Allerede i år øger vi antallet af IT-studiepladser i København med 28 procent. Og vi stopper ikke dér. #uddpol #ITpol #digitalisering #dkbiz
I dag lancerer vi 14 forslag til, hvordan vi løfter de #digitale kompetencer i 🇩🇰. Vi håber, at det bliver vel modtaget af @regeringDK, og at der er politisk vilje til at gøre en indsats for, at 🇩🇰 er klædt på til en kun mere digital fremtid.
altinget.dk/digital/artikel/…
Mads Græsbøll Christensen retweeted
Enig med @marie_bjerre og @Perbbr Vi SKAL gøre noget seriøst ved manglen på IT-specialister. Tårnhøj prioritering på AAU i Aalb. og Kbh, som allerede arbejder intenst på at udvide antallet af IT-studiepladser – uden at vente på penge fra #DKPol
#uddpol #dkbiz @ChristinaEgelun
Tak til @marie_bjerre for at melde klart ud og understrege behovet for at gøre noget ved manglen på it-specialister i Danmark.
På @ITUkbh uddanner vi højt efterspurgte kandidater og kunne skrue markant op med politisk opbakning. #uddpol #itpol
itwatch.dk/ITNyt/Politik/art…
Mads Græsbøll Christensen retweeted
First time we gathered all the research group leaders for two days - it was so rewarding exchanging experiences 🙏
#dkforsk #research @aaucreate
Lyt med på podcasten Athene af Folkeuniversitetet Aalborg hvor jeg fortæller om Fem ting du ikke vidste man kunne med lyd! tinyurl.com/y9tu9r9n
Aktuel Naturvidenskab har skrevet en artikel om Audio Analysis Labs forskning inden for diagnose af Parkinsons ud fra stemmen!
aktuelnaturvidenskab.dk/find…
We have an open postdoc position for our project entitled Edge AI for High Stability-Robustness of Power Electronics in Modern Power Systems which is funded by the Villum Synergy programme for data-driven interdisciplinary research.
stillinger.aau.dk/vis-stilli…
Mads Græsbøll Christensen retweeted
Congrats! This week Audio Analysis Lab celebrated their 10-year anniversary with participants from universities, and companies.
- The lab has strengthened our collaboration with industries working with audio technologies, states founder Professor Græsbøll Christensen, @madsaudio
Mads Græsbøll Christensen retweeted
TILLYKKE med titlen dr.tech. Hør Mads Græsbøll fortælle om sin doktorafhandling - bl.a. om at modellere tale- og lydsignaler. Kan fx bruges til at diagnosticere sygdomme ud fra stemmen og til at forbedre høreapparater. Fascinerende! #audio #dksundhed #dktech
Below is a video presentation of our latest work on sound zones from this year's ICASSP entitled "Robust Pressure matching with ATF Perturbation Constraints for Sound Field Control".
youtu.be/nLSInHdshDQ via @YouTube
4
Here is a presentation from ICASSP of our latest work on distributed privacy-preserving machine learning, more specifically our paper entitled Privacy Preserving Distributed EM for Gaussian Mixture Models using Subspace Perturbation.
youtu.be/q4inSzg_Bx8 via @YouTube
5
Here's the presentation of our work entitled A Bayesian Permutation Training Deep Representation Learning Method for Speech Enhancement from ICASSP. The work is a collaboration between Audio Analysis Lab and Capturi.
youtu.be/ghrwSO7e8JI
1
Here's a presentation of my lab's most recent work on prediction for active noise cancellation in office enviroments youtu.be/eAh0INXvYXg via @YouTube
Below is the 30 minutes presentation of my dr.techn. thesis from my defense on April 29. The thesis is about my work over the past 15+ years on speech and audio modeling. Spoiler alert: the degree was awarded!
youtu.be/HV3VUxie4zI
Below is the 30 minutes presentation of my dr.techn. thesis from my defense on April 29. The thesis is about my work over the past 15+ years on speech and audio modeling. Spoiler alert: the degree was awarded!
youtu.be/HV3VUxie4zI
5
As always, the Audio Analysis Lab is well-represented with papers at ICASSP 2022 (2022.ieeeicassp.org/) which will take place virtually May 7-13 and physically May 22-27, 2022. This year, we have papers on sound zones, ANC, privacy, RIR modeling, and VAEs for enhancement.
8
Gammatone filterbank+simple deep neural network improves the quality of speech for different unseen conditions. The GTFB provides finer resolution in low-frequency regions. Significant improvement of intelligibility and quality for unseen speakers. sciencedirect.com/science/ar…
3
Mads Græsbøll Christensen retweeted
Our most recent work on sound zone control is now avaiable in @ASA_JASA! Check it out.
asa.scitation.org/doi/10.112…









