Unifying Probabilistic Models for Time-Frequency Analysis
| Autor: | Michael Riis Andersen, Dan Stowell, Arno Solin, William J. Wilkinson, Joshua D. Reiss |
|---|---|
| Přispěvatelé: | Queen Mary University of London, Probabilistic Machine Learning, Professorship Solin A., Department of Computer Science, Aalto-yliopisto, Aalto University |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
Signal Processing (eess.SP)
FOS: Computer and information sciences Sound (cs.SD) Computer Science - Machine Learning Computer science Gaussian processes Inference Machine Learning (stat.ML) 02 engineering and technology computer.software_genre Computer Science - Sound Machine Learning (cs.LG) Data modeling symbols.namesake Audio and Speech Processing (eess.AS) Statistics - Machine Learning FOS: Electrical engineering electronic engineering information engineering 0202 electrical engineering electronic engineering information engineering Time domain probabilistic time-frequency analysis Electrical Engineering and Systems Science - Signal Processing Audio signal processing Gaussian process state space models State-space representation Probabilistic logic 020206 networking & telecommunications Amplitude Kernel (statistics) Frequency domain symbols computer Algorithm Electrical Engineering and Systems Science - Audio and Speech Processing |
| Zdroj: | ICASSP |
| Popis: | In audio signal processing, probabilistic time-frequency models have many benefits over their non-probabilistic counterparts. They adapt to the incoming signal, quantify uncertainty, and measure correlation between the signal's amplitude and phase information, making time domain resynthesis straightforward. However, these models are still not widely used since they come at a high computational cost, and because they are formulated in such a way that it can be difficult to interpret all the modelling assumptions. By showing their equivalence to Spectral Mixture Gaussian processes, we illuminate the underlying model assumptions and provide a general framework for constructing more complex models that better approximate real-world signals. Our interpretation makes it intuitive to inspect, compare, and alter the models since all prior knowledge is encoded in the Gaussian process kernel functions. We utilise a state space representation to perform efficient inference via Kalman smoothing, and we demonstrate how our interpretation allows for efficient parameter learning in the frequency domain. Accepted to International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2019 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|