Kevin Wilkinghoff



Kevin Wilkinghoff

I am a postdoctoral researcher at Aalborg University with the Pioneer Centre for AI and an active member of the DCASE community.

My research focuses on anomaly and out-of-distribution detection for audio data, with applications such as acoustic machine condition monitoring, keyword spotting and sound event detection.





Resume: (last update: 07/2025) 

Full CV: (last update: 07/2025) 


Brief CV

02/2025 - present Postdoctoral Researcher
Aalborg University with Pioneer Centre for AI, Denmark
07/2024 - 01/2025 Visiting Research Scientist
MERL, Cambridge, MA, USA
10/2022 - 09/2024 Ph.D. (Dr.rer.nat.) in Computer Science
University of Bonn, Germany
06/2017 - 07/2024 Research Associate
Fraunhofer FKIE, Wachtberg, Germany
10/2014 - 05/2017 M.Sc. in Computer Science
University of Bonn, Germany
10/2010 - 02/2014 B.Sc. in Mathematics
University of Münster, Germany




You can also find me on Google Scholar and dblp.


Preprints

ASDKit: A Toolkit for Comprehensive Evaluation of Anomalous Sound Detection Methods
T. Fujimura, K. Wilkinghoff, K. Imoto, T. Toda
under review

Personalized Speech Synthesis for Zero-Shot Keyword Spotting
F. Gökgöz, A. Cornaggia-Urrigshardt, K. Wilkinghoff
under review

Local Density-Based Anomaly Score Normalization for Domain Generalization
K. Wilkinghoff, H. Yang, J. Ebbers, F. G. Germain, G. Wichern, J. Le Roux
under review

Handling Domain Shifts for Anomalous Sound Detection: A Review of DCASE-Related Work
K. Wilkinghoff, T. Fujimura, K. Imoto, J. Le Roux, Z.-H. Tan, T. Toda
arXiv:2503.10435


Articles in Peer-Reviewed Journals

2024

Why do Angular Margin Losses Work Well for Semi-Supervised Anomalous Sound Detection?
K. Wilkinghoff, F. Kurth
IEEE ACM Trans. Audio Speech Lang. Process.


Articles in Peer-Reviewed Conference and Workshop Proceedings

2025

No Class Left Behind: A Closer Look at Class Balancing for Audio Tagging
J. Ebbers, F. G. Germain, K. Wilkinghoff, G. Wichern, J. Le Roux
ICASSP

Keeping the Balance: Anomaly Score Calculation for Domain Generalization
K. Wilkinghoff, H. Yang, J. Ebbers, F. G. Germain, G. Wichern, J. Le Roux
ICASSP

2024

F1-EV Score: Measuring the Likelihood of Estimating a Good Decision Threshold for Semi-Supervised Anomaly Detection
K. Wilkinghoff, K. Imoto
ICASSP

Self-Supervised Learning for Anomalous Sound Detection
K. Wilkinghoff
ICASSP

TACos: Learning Temporally Structured Embeddings for Few-Shot Keyword Spotting with Dynamic Time Warping
K. Wilkinghoff, A. Cornaggia-Urrigshardt
ICASSP

Multi-Sample Dynamic Time Warping for Few-Shot Keyword Spotting
K. Wilkinghoff, A. Cornaggia-Urrigshardt
EUSIPCO

AdaProj: Adaptively Scaled Angular Margin Subspace Projections for Anomalous Sound Detection with Auxiliary Classification Tasks
K. Wilkinghoff
DCASE

Analyzing the Impact of HF-Specific Signal Degradation on Automatic Speech Recognition
F. Fritz, A. Cornaggia-Urrigshardt, L. Henneke, F. Kurth, K. Wilkinghoff
ICMCIS

Strong Label Generation for Preparing Speech Data in Military Applications Using CTC Loss
F. Gökgöz, A. Cornaggia-Urrigshardt, K. Wilkinghoff
ICMCIS

2023

Design Choices for Learning Embeddings from Auxiliary Tasks for Domain Generalization in Anomalous Sound Detection
K. Wilkinghoff
ICASSP

Novel Generative Classifier for Acoustic Events
P. M. Baggenstoss, K. Wilkinghoff
EUSIPCO

On Using Pre-Trained Embeddings for Detecting Anomalous Sounds with Limited Training Data
K. Wilkinghoff, F. Fritz
EUSIPCO

Language Recognition for SSB modulated HF Radio Signals of Short Duration
A. Cornaggia-Urrigshardt, F. Fritz, L. Henneke, F. Kurth, C. Schlich, K. Wilkinghoff
ITG Speech

2022

Towards Human-Machine Integration for Signal Intelligence Applications
J. D. Rockbach, L.-F. Bluhm, I. Schlangen, L. Over, S. Apfeld, L. Henneke, K. Wilkinghoff
SDF

SCALA-Speech: An Interactive System for Finding and Analyzing Speech Content in Audio Data
A. Cornaggia-Urrigshardt, N. Jarocky, F. Kurth, S. Urrigshardt, K. Wilkinghoff
GI-Jahrestagung

Speech Recognition Lab
A. Cornaggia-Urrigshardt, F. Gökgöz, F. Kurth, H.-C. Schmitz, K. Wilkinghoff
ICMCIS

2021

Sub-Cluster AdaCos: Learning Representations for Anomalous Sound Detection
K. Wilkinghoff
IJCNN

Combining Multiple Distributions based on Sub-Cluster AdaCos for Anomalous Sound Detection under Domain Shifted Conditions
K. Wilkinghoff
DCASE
Best Paper Award

Two-Dimensional Embeddings for Low-Resource Keyword Spotting Based on Dynamic Time Warping
K. Wilkinghoff, A. Cornaggia-Urrigshardt, F. Gökgöz
ITG Speech

2020

On Open-Set Classification with L3-Net Embeddings for Machine Listening Applications
K. Wilkinghoff
EUSIPCO

Using Look, Listen, and Learn Embeddings for Detecting Anomalous Sounds in Machine Condition Monitoring
K. Wilkinghoff
DCASE

On Open-Set Speaker Identification with I-Vectors
K. Wilkinghoff
Odyssey

Towards Robust Speech Interfaces for the ISS
H.-C. Schmitz, F. Kurth, K. Wilkinghoff, U. Müllerschkowski, C. Karrasch, V. Schmid
IUI Companion

2019

Open-Set Acoustic Scene Classification with Deep Convolutional Autoencoders
K. Wilkinghoff, F. Kurth
DCASE

Calm Interfaces for Integrated C2 Systems
H.-C. Schmitz, A. Cornaggia-Urrigshardt, F. Gökgöz, S. Kent, K. Wilkinghoff
ICCRTS

2018

Robust Detection of Jittered Multiply Repeating Audio Events Using Iterated Time-Warped ACF
F. Kurth, K. Wilkinghoff
ICASSP

General-Purpose Audio Tagging by Ensembling Convolutional Neural Networks based on Multiple Features
K. Wilkinghoff
DCASE

Accurately Capturing Speech Feature Distributions by Extending Supervectors for Robust Speaker Recognition
K. Wilkinghoff
ITG Speech

Robust Speaker Identification by Fusing Classification Scores with a Neural Network
K. Wilkinghoff, P. M. Baggenstoss, A. Cornaggia-Urrigshardt, F. Kurth
ITG Speech

2017

Glottal Mixture Model (GLOMM) for Speaker Identification on Telephone Channels
P. M. Baggenstoss, K. Wilkinghoff, F. Kurth
EUSIPCO


Invited Articles in Conference Proceedings without Peer-Review

2025

Handling Domain Shifts for Anomalous Sound Detection: A Review
K. Wilkinghoff, T. Fujimura, K. Imoto, J. Le Roux
DAS/DAGA

2022

On choosing decision thresholds for anomalous sound detection in machine condition monitoring
K. Wilkinghoff, A. Cornaggia-Urrigshardt
ICA


Technical Reports for Academic Challenges

2024

FKIE-VUB System for DCASE2024 Challenge Task 2: First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring
K. Wilkinghoff, Y. Bel-Hadj
DCASE2024 Challenge, Ranked 19/28.

2023

Fraunhofer FKIE Submission for Task 2: First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring
K. Wilkinghoff
DCASE2023 Challenge, Ranked 4/24

Fraunhofer FKIE Submission for Task 5: Few-Shot Bioacoustic Event Detection
K. Wilkinghoff, A. Cornaggia-Urrigshardt
DCASE2023 Challenge, Ranked 6/8

2022

An Outlier Exposed Anomalous Sound Detection System for Domain Generalization in Machine Condition Monitoring
K. Wilkinghoff
DCASE2022 Challenge, Ranked 10/32

2021

Utilizing Sub-Cluster Adacos for Anomalous Sound Detection under Domain Shifted Conditions
K. Wilkinghoff
DCASE2021 Challenge, Ranked 3/27
Judges' Award

2020

Anomalous Sound Detection with Look, Listen, and Learn Embeddings
K. Wilkinghoff
DCASE2020 Challenge, Ranked 8/40

2019

Open-Set Acoustic Scene Classification with Deep Convolutional Autoencoders
K. Wilkinghoff, F. Kurth
DCASE2019 Challenge, Ranked 19/39 and 3/7

2018

Open-Set Speaker Recognition with Augmented i-Vectors
K. Wilkinghoff
MCE2018 Challenge, Ranked 8/13 and 6/13

General-Purpose Audio Tagging by Ensembling Convolutional Neural Networks based on Multiple Features
K. Wilkinghoff
DCASE2018 Challenge, Ranked 14/39


Theses

2024

Audio Embeddings for Semi-Supervised Anomalous Sound Detection
Doctoral Thesis, Faculty of Mathematics and Natural Sciences, University of Bonn

2017

Neural Networks in Speaker Identification
Master's Thesis, Institute of Computer Science, University of Bonn
Kölner VDI Förderpreis 2018
AFCEA Bonn Studienpreis 2018

2013

Homogenisierung nichtlinearer elliptischer Differentialgleichungen
Bachelor's Thesis, Mathematical Institute, University of Münster