Kevin Wilkinghoff
![Kevin Wilkinghoff](/public/img/photography/people/wilkinghoff.jpg)
- Phone: 617-621-7551
- Email:
-
Position:
Research / Technical Staff
Visiting Research Scientist -
Education:
M.Sc., University of Bonn, 2017 -
Research Area:
Kevin's Quick Links
-
Biography
Before joining MERL, Kevin was a research associate with Fraunhofer FKIE and completed his Ph.D. studies in computer science at the University of Bonn. His current research focuses on anomaly detection and representation learning in the context of sound event detection. In 2021, he was the recipient of the DCASE Best Paper Award and a DCASE Challenge Judge's Award for his work on anomalous sound detection.
-
Other Publications
- "Analyzing the Impact of HF-Specific Signal Degradation on Automatic Speech Recognition", International Conference on Military Communications and Information Systems (ICMCIS), 2024.BibTeX
- @Inproceedings{fritz2024analyzing,
- author = {Fritz, Fabian and Cornaggia-Urrigshardt, Alessia and Henneke, Lukas and Kurth, Frank and Wilkinghoff, Kevin},
- title = {Analyzing the Impact of HF-Specific Signal Degradation on Automatic Speech Recognition},
- booktitle = {International Conference on Military Communications and Information Systems (ICMCIS)},
- year = 2024,
- publisher = {IEEE}
- }
, - "Strong Label Generation for Preparing Speech Data in Military Applications Using CTC Loss", International Conference on Military Communications and Information Systems (ICMCIS), 2024.BibTeX
- @Inproceedings{goekgoez2024strong,
- author = {G\"okg\"oz, Fahrettin and Cornaggia-Urrigshardt, Alessia and Wilkinghoff, Kevin},
- title = {Strong Label Generation for Preparing Speech Data in Military Applications Using CTC Loss},
- booktitle = {International Conference on Military Communications and Information Systems (ICMCIS)},
- year = 2024,
- publisher = {IEEE}
- }
, - "F1-EV Score: Measuring the Likelihood of Estimating a Good Decision Threshold for Semi-Supervised Anomaly Detection", International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024, pp. 256-260.BibTeX
- @Inproceedings{wilkinghoff2024f1-ev,
- author = {Wilkinghoff, Kevin and Imoto, Keisuke},
- title = {F1-EV Score: Measuring the Likelihood of Estimating a Good Decision Threshold for Semi-Supervised Anomaly Detection},
- booktitle = {International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
- year = 2024,
- pages = {256--260},
- publisher = {IEEE}
- }
, - "FKIE-VUB System for DCASE2024 Challenge Task 2: First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring," Tech. Rep., DCASE2024 Challenge, 2024.BibTeX
- @Techreport{wilkinghoff2024fkie-vub,
- author = {Wilkinghoff, Kevin and Bel-Hadj, Yacine},
- title = {FKIE-VUB System for DCASE2024 Challenge Task 2: First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring},
- institution = {DCASE2024 Challenge},
- year = 2024
- }
, - "Self-Supervised Learning for Anomalous Sound Detection", International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024, pp. 276-280.BibTeX
- @Inproceedings{wilkinghoff2024self,
- author = {Wilkinghoff, Kevin},
- title = {Self-Supervised Learning for Anomalous Sound Detection},
- booktitle = {International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
- year = 2024,
- pages = {276--280},
- publisher = {IEEE}
- }
, - "TACos: Learning Temporally Structured Embeddings for Few-Shot Keyword Spotting with Dynamic Time Warping", International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024, pp. 9941-9945.BibTeX
- @Inproceedings{wilkinghoff2024tacos,
- author = {Wilkinghoff, Kevin and Cornaggia-Urrigshardt, Alessia},
- title = {TACos: Learning Temporally Structured Embeddings for Few-Shot Keyword Spotting with Dynamic Time Warping},
- booktitle = {International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
- year = 2024,
- pages = {9941--9945},
- publisher = {IEEE}
- }
, - "Why do Angular Margin Losses work well for Semi-Supervised Anomalous Sound Detection?", IEEE/ACM Transactions on Audio, Speech and Language Processing (TASLP), Vol. 32, pp. 608-622, 2024.BibTeX
- @Article{wilkinghoff2024why,
- author = {Wilkinghoff, Kevin and Kurth, Frank},
- title = {Why do Angular Margin Losses work well for Semi-Supervised Anomalous Sound Detection?},
- journal = {IEEE/ACM Transactions on Audio, Speech and Language Processing (TASLP)},
- year = 2024,
- volume = 32,
- pages = {608--622}
- }
, - "Fraunhofer FKIE Submission for Task 5: Few-Shot Bioacoustic Event Detection," Tech. Rep., DCASE2023 Challenge, 2023.BibTeX
- @Techreport{Wilkinghoff2023fraunhofer5-challenge,
- author = {Wilkinghoff, Kevin and Cornaggia-Urrigshardt, Alessia},
- title = {Fraunhofer FKIE Submission for Task 5: Few-Shot Bioacoustic Event Detection},
- institution = {DCASE2023 Challenge},
- year = 2023
- }
, - "Novel Generative Classifier for Acoustic Events", 31st European Signal Processing Conference (EUSIPCO), 2023, pp. 196-200.BibTeX
- @Inproceedings{baggenstoss2023novel,
- author = {Baggenstoss, Paul M. and Wilkinghoff, Kevin},
- title = {Novel Generative Classifier for Acoustic Events},
- booktitle = {31st European Signal Processing Conference (EUSIPCO)},
- year = 2023,
- pages = {196--200},
- publisher = {IEEE}
- }
, - "Language Recognition for SSB modulated HF Radio Signals of Short Duration", 15th ITG Conference on Speech Communication (ITG Speech), 2023, pp. 240-244.BibTeX
- @Inproceedings{cornaggia2023language,
- author = {Cornaggia-Urrigshardt, Alessia and Fritz, Fabian and Henneke, Lukas and Kurth, Frank and Schlich, Christian and Wilkinghoff, Kevin},
- title = {Language Recognition for SSB modulated HF Radio Signals of Short Duration},
- booktitle = {15th ITG Conference on Speech Communication (ITG Speech)},
- year = 2023,
- pages = {240--244},
- publisher = {VDE}
- }
, - "Design Choices for Learning Embeddings from Auxiliary Tasks for Domain Generalization in Anomalous Sound Detection", International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023.BibTeX
- @Inproceedings{wilkinghoff2023design,
- author = {Wilkinghoff, Kevin},
- title = {Design Choices for Learning Embeddings from Auxiliary Tasks for Domain Generalization in Anomalous Sound Detection},
- booktitle = {International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
- year = 2023,
- publisher = {IEEE}
- }
, - "Fraunhofer FKIE Submission for Task 2: First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring," Tech. Rep., DCASE2023 Challenge, 2023.BibTeX
- @Techreport{wilkinghoff2023fraunhofer2-challenge,
- author = {Wilkinghoff, Kevin},
- title = {Fraunhofer FKIE Submission for Task 2: First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring},
- institution = {DCASE2023 Challenge},
- year = 2023
- }
, - "On Using Pre-Trained Embeddings for Detecting Anomalous Sounds with Limited Training Data", 31st European Signal Processing Conference (EUSIPCO), 2023, pp. 186-190.BibTeX
- @Inproceedings{wilkinghoff2023using,
- author = {Wilkinghoff, Kevin and Fritz, Fabian},
- title = {On Using Pre-Trained Embeddings for Detecting Anomalous Sounds with Limited Training Data},
- booktitle = {31st European Signal Processing Conference (EUSIPCO)},
- year = 2023,
- pages = {186--190},
- publisher = {IEEE}
- }
, - "SCALA-Speech: An Interactive System for Finding and Analyzing Speech Content in Audio Data", 52. Jahrestagung der Gesellschaft für Informatik (GI-Jahrestagung), 2022, vol. {P-326}, pp. 81-90.BibTeX
- @Inproceedings{cornaggia2022scala,
- author = {Cornaggia-Urrigshardt, Alessia and Jarocky, Nikita and Kurth, Frank and Urrigshardt, Sebastian and Wilkinghoff, Kevin},
- title = {SCALA-Speech: An Interactive System for Finding and Analyzing Speech Content in Audio Data},
- booktitle = {52. Jahrestagung der Gesellschaft f{\"{u}}r Informatik (GI-Jahrestagung)},
- year = 2022,
- volume = {P-326},
- series = {LNI},
- pages = {81--90},
- publisher = {GI}
- }
, - "Speech Recognition Lab", International Conference on Military Communications and Information Systems (ICMCIS), 2022, vol. 205, pp. 218-228.BibTeX
- @Inproceedings{cornaggia2022speech,
- author = {Cornaggia-Urrigshardt, Alessia and G\"{o}kg\"{o}z, Fahrettin and Kurth, Frank and Schmitz, Hans-Christian and Wilkinghoff, Kevin},
- title = {Speech Recognition Lab},
- booktitle = {International Conference on Military Communications and Information Systems (ICMCIS)},
- year = 2022,
- volume = 205,
- series = {Procedia Computer Science},
- pages = {218--228},
- publisher = {Elsevier}
- }
, - "Towards Human-Machine Integration for Signal Intelligence Applications", 14th Symposium Sensor Data Fusion (SDF), 2022.BibTeX
- @Inproceedings{rockbach2022human,
- author = {Rockbach, Jonas D. and Bluhm, Luka-Franziska and Schlangen, Isabel and Over, Laura and Apfeld, Sabine and Henneke, Lukas and Wilkinghoff, Kevin},
- title = {Towards Human-Machine Integration for Signal Intelligence Applications},
- booktitle = {14th Symposium Sensor Data Fusion (SDF)},
- year = 2022,
- publisher = {IEEE}
- }
, - "On choosing decision thresholds for anomalous sound detection in machine condition monitoring", 24th International Congress on Acoustics (ICA), 2022.BibTeX
- @Inproceedings{wilkinghoff2022choosing,
- author = {Wilkinghoff, Kevin and Cornaggia-Urrigshardt, Alessia},
- title = {On choosing decision thresholds for anomalous sound detection in machine condition monitoring},
- booktitle = {24th International Congress on Acoustics (ICA)},
- year = 2022,
- publisher = {The Acoustical Society of Korea}
- }
, - "An Outlier Exposed Anomalous Sound Detection System for Domain Generalization in Machine Condition Monitoring," Tech. Rep., DCASE2022 Challenge, 2022.BibTeX
- @Techreport{wilkinghoff2022outlier-challenge,
- author = {Wilkinghoff, Kevin},
- title = {An Outlier Exposed Anomalous Sound Detection System for Domain Generalization in Machine Condition Monitoring},
- institution = {DCASE2022 Challenge},
- year = 2022
- }
, - "Combining Multiple Distributions based on Sub-Cluster AdaCos for Anomalous Sound Detection under Domain Shifted Conditions", Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE), 2021, pp. 55-59.BibTeX
- @Inproceedings{wilkinghoff2021combining,
- author = {Wilkinghoff, Kevin},
- title = {Combining Multiple Distributions based on Sub-Cluster AdaCos for Anomalous Sound Detection under Domain Shifted Conditions},
- booktitle = {Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE)},
- year = 2021,
- pages = {55--59}
- }
, - "Sub-Cluster AdaCos: Learning Representations for Anomalous Sound Detection", International Joint Conference on Neural Networks (IJCNN), 2021.BibTeX
- @Inproceedings{wilkinghoff2021sub,
- author = {Wilkinghoff, Kevin},
- title = {Sub-Cluster AdaCos: Learning Representations for Anomalous Sound Detection},
- booktitle = {International Joint Conference on Neural Networks (IJCNN)},
- year = 2021,
- publisher = {IEEE}
- }
, - "Two-Dimensional Embeddings for Low-Resource Keyword Spotting Based on Dynamic Time Warping", 14th ITG Conference on Speech Communication (ITG Speech), 2021, pp. 9-13.BibTeX
- @Inproceedings{wilkinghoff2021twodimensional,
- author = {Wilkinghoff, Kevin and Cornaggia-Urrigshardt, Alessia and G\"{o}kg\"{o}z, Fahrettin},
- title = {Two-Dimensional Embeddings for Low-Resource Keyword Spotting Based on Dynamic Time Warping},
- booktitle = {14th ITG Conference on Speech Communication (ITG Speech)},
- year = 2021,
- pages = {9--13},
- publisher = {VDE}
- }
, - "Utilizing Sub-Cluster Adacos for Anomalous Sound Detection under Domain Shifted Conditions," Tech. Rep., DCASE2021 Challenge, 2021.BibTeX
- @Techreport{wilkinghoff2021utilizing-challenge,
- author = {Wilkinghoff, Kevin},
- title = {Utilizing Sub-Cluster Adacos for Anomalous Sound Detection under Domain Shifted Conditions},
- institution = {DCASE2021 Challenge},
- year = 2021
- }
, - "Towards Robust Speech Interfaces for the ISS", International Conference on Intelligent User Interfaces (IUI) Companion, 2020, pp. 110-111.BibTeX
- @Inproceedings{schmitz2020towards,
- author = {Schmitz, Hans-Christian and Kurth, Frank and Wilkinghoff, Kevin and M\"{u}llerschkowski, Uwe and Karrasch, Christian and Schmid, Volker},
- title = {Towards Robust Speech Interfaces for the ISS},
- booktitle = {International Conference on Intelligent User Interfaces (IUI) Companion},
- year = 2020,
- pages = {110--111},
- publisher = {ACM}
- }
, - "Using Look, Listen, and Learn Embeddings for Detecting Anomalous Sounds in Machine Condition Monitoring", Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE), 2020, pp. 215-219.BibTeX
- @Inproceedings{wilkinghoff2020anomalous,
- author = {Wilkinghoff, Kevin},
- title = {Using Look, Listen, and Learn Embeddings for Detecting Anomalous Sounds in Machine Condition Monitoring},
- booktitle = {Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE)},
- year = 2020,
- pages = {215--219}
- }
, - "Anomalous Sound Detection with Look, Listen, and Learn Embeddings," Tech. Rep., DCASE2020 Challenge, 2020.BibTeX
- @Techreport{wilkinghoff2020anomalous-challenge,
- author = {Wilkinghoff, Kevin},
- title = {Anomalous Sound Detection with Look, Listen, and Learn Embeddings},
- institution = {DCASE2020 Challenge},
- year = 2020,
- note = {Ranked 8/40}
- }
, - "On Open-Set Classification with L3-Net Embeddings for Machine Listening Applications", 28th European Signal Processing Conference (EUSIPCO), 2020, pp. 800-804.BibTeX
- @Inproceedings{wilkinghoff2020open,
- author = {Wilkinghoff, Kevin},
- title = {On Open-Set Classification with L3-Net Embeddings for Machine Listening Applications},
- booktitle = {28th European Signal Processing Conference (EUSIPCO)},
- year = 2020,
- pages = {800--804},
- publisher = {IEEE}
- }
, - "On Open-Set Speaker Identification with I-Vectors", Odyssey - The Speaker and Language Recognition Workshop (Odyssey), 2020, pp. 408-414.BibTeX
- @Inproceedings{wilkinghoff2020openspeaker,
- author = {Wilkinghoff, Kevin},
- title = {On Open-Set Speaker Identification with I-Vectors},
- booktitle = {Odyssey - The Speaker and Language Recognition Workshop (Odyssey)},
- year = 2020,
- pages = {408--414},
- publisher = {ISCA}
- }
, - "Calm Interfaces for Integrated C2 Systems", International Command and Control Research and Technology Symposium (ICCRTS), 2019.BibTeX
- @Inproceedings{schmitz2019calm,
- author = {Schmitz, Hans-Christian and Cornaggia-Urrigshardt, Alessia and G\"{o}kg\"{o}z, Fahrettin and Kent, Samantha and Wilkinghoff, Kevin},
- title = {Calm Interfaces for Integrated C2 Systems},
- booktitle = {International Command and Control Research and Technology Symposium (ICCRTS)},
- year = 2019
- }
, - "Open-Set Acoustic Scene Classification with Deep Convolutional Autoencoders", Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE), 2019, pp. 258-262.BibTeX
- @Inproceedings{wilkinghoff2019open,
- author = {Wilkinghoff, Kevin and Kurth, Frank},
- title = {Open-Set Acoustic Scene Classification with Deep Convolutional Autoencoders},
- booktitle = {Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE)},
- year = 2019,
- pages = {258--262},
- publisher = {New York University}
- }
, - "Open-Set Acoustic Scene Classification with Deep Convolutional Autoencoders," Tech. Rep., DCASE2019 Challenge, 2019.BibTeX
- @Techreport{wilkinghoff2019open-challenge,
- author = {Wilkinghoff, Kevin and Kurth, Frank},
- title = {Open-Set Acoustic Scene Classification with Deep Convolutional Autoencoders},
- institution = {DCASE2019 Challenge},
- year = 2019
- }
, - "Robust Detection of Jittered Multiply Repeating Audio Events using Iterated Time-Warped ACF", International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018, pp. 276-280.BibTeX
- @Inproceedings{kurth2018twacf,
- author = {Kurth, Frank and Wilkinghoff, Kevin},
- title = {Robust Detection of Jittered Multiply Repeating Audio Events using Iterated Time-Warped ACF},
- booktitle = {International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
- year = 2018,
- pages = {276--280},
- publisher = {IEEE}
- }
, - "General-Purpose Audio Tagging by Ensembling Convolutional Neural Networks based on Multiple Features", Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE), 2018, pp. 44-48.BibTeX
- @Inproceedings{wilkinghoff2018dcase,
- author = {Wilkinghoff, Kevin},
- title = {General-Purpose Audio Tagging by Ensembling Convolutional Neural Networks based on Multiple Features},
- booktitle = {Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE)},
- year = 2018,
- pages = {44--48},
- publisher = {Tampere University of Technology}
- }
, - "Accurately Capturing Speech Feature Distributions by Extending Supervectors for Robust Speaker Recognition", 13th ITG Symposium on Speech Communication (ITG Speech), 2018, pp. 81-85.BibTeX
- @Inproceedings{wilkinghoff2018extsv,
- author = {Wilkinghoff, Kevin},
- title = {Accurately Capturing Speech Feature Distributions by Extending Supervectors for Robust Speaker Recognition},
- booktitle = {13th ITG Symposium on Speech Communication (ITG Speech)},
- year = 2018,
- pages = {81--85},
- publisher = {VDE}
- }
, - "General-Purpose Audio Tagging by Ensembling Convolutional Neural Networks based on Multiple Features," Tech. Rep., DCASE2018 Challenge, 2018.BibTeX
- @Techreport{wilkinghoff2018general-challenge,
- author = {Wilkinghoff, Kevin},
- title = {General-Purpose Audio Tagging by Ensembling Convolutional Neural Networks based on Multiple Features},
- institution = {DCASE2018 Challenge},
- year = 2018
- }
, - "Robust Speaker Identification by Fusing Classification Scores with a Neural Network", 13th ITG Symposium on Speech Communication (ITG Speech), 2018, pp. 261-265.BibTeX
- @Inproceedings{wilkinghoff2018neuralfusion,
- author = {Wilkinghoff, Kevin and Baggenstoss, Paul M. and Cornaggia-Urrigshardt, Alessia and Kurth, Frank},
- title = {Robust Speaker Identification by Fusing Classification Scores with a Neural Network},
- booktitle = {13th ITG Symposium on Speech Communication (ITG Speech)},
- year = 2018,
- pages = {261--265},
- publisher = {VDE}
- }
, - "Open-Set Speaker Recognition with Augmented i-Vectors," Tech. Rep., MCE2018 Challenge, 2018.BibTeX
- @Techreport{wilkinghoff2018open-challenge,
- author = {Wilkinghoff, Kevin},
- title = {Open-Set Speaker Recognition with Augmented i-Vectors},
- institution = {MCE2018 Challenge},
- year = 2018
- }
, - "Glottal Mixture Model (GLOMM) for Speaker Identification on Telephone Channels", 25th European Signal Processing Conference (EUSIPCO), 2017, pp. 2734-2738.BibTeX
- @Inproceedings{baggenstoss2017glomm,
- author = {Baggenstoss, Paul M. and Wilkinghoff, Kevin and Kurth, Frank},
- title = {Glottal Mixture Model (GLOMM) for Speaker Identification on Telephone Channels},
- booktitle = {25th European Signal Processing Conference (EUSIPCO)},
- year = 2017,
- pages = {2734--2738},
- publisher = {IEEE}
- }
,
- "Analyzing the Impact of HF-Specific Signal Degradation on Automatic Speech Recognition", International Conference on Military Communications and Information Systems (ICMCIS), 2024.