Niels Henrik Pontoppidan — accepted AI-conference papers

2019

Machine Learning for Condition Monitoring and Innovation

Niels Henrik Pontoppidan, Tue Lehn-Schiøler, Kaare Brandt Petersen

ICASSP 2019accepted

This paper is a tribute to the late Professor Jan Larsen's work on machine learning for condition monitoring of large diesel engines. We present the ultrasound signals used to monitor the condition of the large engines and revisit two methods for estimating the condition of the engines from the ultr…

Cited by 0 Source ↗Scholar ↗

2015

Low-latency sound-source-separation using non-negative matrix factorisation with coupled analysis and synthesis dictionaries

Tom Barker, Tuomas Virtanen, Niels Henrik Pontoppidan

ICASSP 2015accepted

For real-time or close to real-time applications, sound source separation can be performed on-line, where new frames of incoming data for a mixture signal are processed as they arrive, at very low delay. We propose an approach which generates the separation filters for short synthesis frames to achi…

Cited by 0 Source ↗Scholar ↗