Nilesh Madhu

NileshMadhu.jpg

Nilesh Madhu received his Dr.-Ing. degree (summa cum laude) from the Ruhr-Universität Bochum, Germany, in 2009. His dissertation was on signal processing algorithms for the localisation and separation of acoustic sources using microphone arrays. Following this he was awarded a Marie-Curie experienced researcher fellowship for a two-year postdoctoral stay at the KU Leuven, Belgium. Here he successfully applied his signal processing knowledge to the field of hearing prostheses and biomedical signal analysis.

From 2011 to 2017 he was with NXP Semiconductors, Belgium, where he held the positions of principal scientist and team lead within the product line Mobile Audio Solutions. During this period he and his team were tasked with developing innovative, beyond the state-of-the-art algorithms for audio and speech enhancement for mobile communications devices. This work, consistently held to exacting industry standards and strict delivery deadlines, led to the successful deployment of their algorithms on several flagship models of major smartphone OEM's.

Since December 2017 he is a professor (tenure track) for audio and speech processing at Ghent University and imec, Belgium. He is passionate about signal processing and is especially interested in signal detection and enhancement for various applications in the fields of communications, healthcare and automation.

Information about the research focus of the audio and speech processing group can be found here.

Publications

Recent publications

  • R. Verhack, N. Madhu, G.V. Wallendael, P. Lambert, T. Sikora, "Steered mixture-of-experts approximation of spherical image data,", in Proceedings of the 26th European Signal Processing Conference (EUSIPCO), pp. 1-5, Sept. 3-7, 2018.
  • S. Elshamy, N. Madhu, W. Tirry, and T. Fingscheidt, "A priori SNR computation for speech enhancement based on cepstral envelope estimation," in Proceedings of the 16th International Workshop on Acoustic Signal Enhancement (IWAENC), pp. 1–5, Sept. 17-20, 2018.
  • S. Gergen, R. Martin, and N. Madhu, "Source separation by feature-based clustering of microphones in ad hoc arrays," in Proceedings of the 16th International Workshop on Acoustic Signal Enhancement (IWAENC), pp. 1–5, 2018, Sept. 17-20, 2018.
  • N. Madhu and R. Martin, "Source number estimation for multi-speaker localisation and tracking," in Proceedings of the Workshop on Speech Processing for Voice, Speech and Hearing Disorders, pp. 1-5, Sept. 8-9, 2018.
  • G. Vanderreydt and N. Madhu, "Analysis of an online single-channel self-adaptive model-based voice activity detector and suggestions for improvement," in Proceedings of the Workshop on Speech Processing for Voice, Speech and Hearing Disorders, (Abstract/Poster), Sept. 8-9, 2018.
  • S. Elshamy, N. Madhu, W. Tirry, and T. Fingscheidt, "DNN-supported speech enhancement with cepstral estimation of both excitation and envelope," IEEE/ACM Trans. Audio, Speech, and Language Processing, vol. 26, no. 12, pp. 2460–2474, 2018.
  • S. Gergen, R. Martin, and N. Madhu, "Source separation by source separation by fuzzy-membership value aware beamforming and masking in ad hoc arrays," in Proceedings of the 13th ITG Conference on Voice Communication, pp. 1–5, Oct. 10-12, 2018.

Contact

Phone +32 488 04 25 14

Email nilesh.madhu@ugent.be

Researcher identification

ORCID: 0000-0001-9131-3309

Research-ID: L-2268-2018