Alexandru Nelus M.Sc.

Ruhr-Universität Bochum
Institut für Kommunikationsakustik
Fakultät für Elektrotechnik und Informationstechnik
Universitätsstr. 150
D-44780 Bochum
Raum: ID/2/221

Email: alexandru.nelus@rub.de
Tel.: +49 234 32 25388

Publications

Nelus, A., Glitza, R., & Martin, R. (2021). Estimation of Microphone Clusters in Acoustic Sensor Networks Using Unsupervised Federated Learning. In ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (pp. 761–765). IEEE. (https://doi.org/10.1109/ICASSP39728.2021.9414186)

Nelus, A., Glitza, R., & Martin, R. (2021). Unsupervised Clustered Federated Learning in Complex Multi-source Acoustic Environments. In 2021 29th European Signal Processing Conference (EUSIPCO) (pp. 1115–1119). IEEE. (https://doi.org/10.23919/EUSIPCO54536.2021.9615980)

Nelus, A., & Martin, R. (2021). Privacy-preserving Audio Classification using Variational Information Feature Extraction. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2864–2877. (https://doi.org/10.1109/TASLP.2021.3108063)

Becker, L., Nelus, A., Gauer, J., Rudolph, L., & Martin, R. (2020). Audio Feature Extraction for Vehicle Engine Noise Classification. In ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 711-715). IEEE. (https://doi.org/10.1109/ICASSP40776.2020.9053117)

Nelus, A., Ebbers, J., Haeb-Umbach, R., & Martin, R. (2019). Privacy-Preserving Variational Information Feature Extraction for Domestic Activity Monitoring versus Speaker Identification. In Interspeech 2019 (pp. 3710–3714). ISCA. (https://doi.org/10.21437/Interspeech.2019-1703)

Nelus, A., Rech, S., Koppelmann, T., Biermann, H., & Martin, R. (2019). Privacy-Preserving Siamese Feature Extraction for Gender Recognition versus Speaker Identification. In Interspeech 2019 (pp. 3705–3709). ISCA. (https://doi.org/10.21437/Interspeech.2019-1148)

Nelus, A., & Martin, R. (2019). Privacy-aware Feature Extraction for Gender Discrimination versus Speaker Identification. In 2019 IEEE International Conference on Acoustics, Speech and Signal Processing: Proceedings : May 12-17, 2019, Brighton Conference Centre, Brighton, United Kingdom (pp. 671–674). IEEE. (https://doi.org/10.1109/ICASSP.2019.8682394)

Ebbers, J., Nelus, A., Martin, R., & Häb-Umbach, R. (2018). Evaluation of modulation-MFCC features and DNN classification for acoustic event detection. In Tagungsband - DAGA 2018: 44. Deutsche Jahrestagung für Akustik : 19.-22. März 2018, München. DEGA.

Nelus, A., & Martin, R. (2018). Gender discrimination versus speaker identification through privacy-aware adversarial feature extraction. In S. Doclo & P. Jax (Eds.), ITG-Fachbericht: Vol. 282. Speech communication: 13. Itg-Fachtagung Sprachkommunikation 10.- 12. Oktober 2018 in Oldenburg (pp. 101–105). VDE VERLAG.

Nelus, A., Gergen, S. & Martin, R. (2017). Analysis of temporal aggregation and dimensionality reduction on feature sets for speaker identification in wireless acoustic sensor networks. In 2017 IEEE 19th International Workshop on Multimedia Signal Processing (MMSP). IEEE. (https://doi.org/10.1109/MMSP.2017.8122277)

Nelus, A., Gergen, S., Taghia, J. & Martin, R. (2016). Towards Opaque Audio Features for Privacy in Acoustic Sensor Networks. In R. Häb-Umbach & S. Doclo (Hrsg.), ITG-Fachbericht: Bd. 267, Speech Communication: 12. ITG-Fachtagung Sprachkommunikation, 5.-7. Oktober 2016 in Paderborn. VDE Verlag.

Nelus, A., Nicolae, M., & Popescu, D. (2016). Simulation framework for wsn used in monitoring of illegal tree cutting. University Politehnica of Bucharest Scientific Bulletin Series C-Electrical Engineering and Computer Science, vol. 78, no. 3, pp. 27–38, 2016.

Awards

Best Student Paper Award at 13th ITG Conference on Speech Communication Oldenburg, October 10-12, 2018. Paper: A. Nelus and R. Martin, “Gender discrimination versus speaker identification through privacy-aware adversarial feature extraction,”

Other

Presentation on “Privacy-preserving Adversarial Feature Extraction in Speaker Classification Tasks” at Data Science Ruhrgebiet 2019.

Organizer of Android @RUB Hackathon 2019, Institut für Kommunikationsakustik, Ruhr-Universität Bochum. https://www.eventbrite.com/e/android-rub-hackathon-2019-tickets-53256492603#

Poster presentation A. Nelus and R. Martin, “Scalable audio features for clustering and classification with privacy constraints,” in Satellite Workshop „Acoustic Sensor Networks“ in Speech Communication; 13. ITG Symposium; 2018.

Organizer of Android @RUB Hackathon 2017, Institut für Kommunikationsakustik, Ruhr-Universität Bochum. https://www.eventbrite.com/e/android-rub-hackathon-tickets-33657751240# 

Coordinator of AppTeam Ruhr University Bochum, https://play.google.com/store/apps/developer?id=AppTeam+Ruhr+University+Bochum.

Teaching

Bachelor-Vertiefungspraktikum Elektrotechnik und Informationstechnik IT-V3  WS15, WS16, WS17, WS18.

Bachelor-Vertiefungsseminar Informationstechnik SS16, SS17, SS18, SS19.

Grundlagen der Sprachsignalverarbeitung  WS16.

Bachelor project supervision:

  • Audio signal labeling and classification on Android based embedded devices.
  • Real-time localization, beamforming and noise reduction, joint supervision with Mehdi Zohourian.
  • Music genre classification using Mod-MFCC features.
  • Gender discrimination vs. speaker identification through privacy-aware siamese feature extraction.
  • Implementation of privacy-preserving feature extraction in a distributed acoustic sensor network.
  • End-to-end approximation of auditory models using artificial neuronal networks, joint supervision with Anil Nagathil.

Bachelor thesis supervision:

  • Audio feature extraction with privacy constraints on Android based embedded devices.
  • Audio signal classification with privacy constraints on Android based embedded devices.
  • Automatic classification of moving vehicles using audio signals, joint supervision with Johannes Gauer.
  • Generation of cryptographic keys using the available information of acoustic channels      .
  • Content- and context-based classification of music signals using deep neural networks.
  • Gender discrimination vs. speaker identification through privacy-aware siamese feature extraction.
  • Analysis of privacy-preserving feature extraction schemes for domestic activity monitoring vs. speaker identification.

Foto Alexandru Nelus