Gabor Frames and Deep Scattering Networks in Audio Processing
This paper introduces Gabor scattering, a feature extractor based on Gabor frames and Mallat’s scattering transform. By using a simple signal model for audio signals, specific properties of Gabor scattering are studied. It is shown that, for each layer, specific invariances to certain sign...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-09-01
|
| Series: | Axioms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1680/8/4/106 |
| _version_ | 1818452695041703936 |
|---|---|
| author | Roswitha Bammer Monika Dörfler Pavol Harar |
| author_facet | Roswitha Bammer Monika Dörfler Pavol Harar |
| author_sort | Roswitha Bammer |
| collection | DOAJ |
| description | This paper introduces Gabor scattering, a feature extractor based on Gabor frames and Mallat’s scattering transform. By using a simple signal model for audio signals, specific properties of Gabor scattering are studied. It is shown that, for each layer, specific invariances to certain signal characteristics occur. Furthermore, deformation stability of the coefficient vector generated by the feature extractor is derived by using a decoupling technique which exploits the contractivity of general scattering networks. Deformations are introduced as changes in spectral shape and frequency modulation. The theoretical results are illustrated by numerical examples and experiments. Numerical evidence is given by evaluation on a synthetic and a “real” dataset, that the invariances encoded by the Gabor scattering transform lead to higher performance in comparison with just using Gabor transform, especially when few training samples are available. |
| first_indexed | 2024-12-14T21:27:10Z |
| format | Article |
| id | doaj.art-4f55347adfa2428cabfb0cc1a8c0c643 |
| institution | Directory Open Access Journal |
| issn | 2075-1680 |
| language | English |
| last_indexed | 2024-12-14T21:27:10Z |
| publishDate | 2019-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Axioms |
| spelling | doaj.art-4f55347adfa2428cabfb0cc1a8c0c6432022-12-21T22:46:47ZengMDPI AGAxioms2075-16802019-09-018410610.3390/axioms8040106axioms8040106Gabor Frames and Deep Scattering Networks in Audio ProcessingRoswitha Bammer0Monika Dörfler1Pavol Harar2NuHAG, Faculty of Mathematics, University of Vienna, 1090 Wien, AustriaNuHAG, Faculty of Mathematics, University of Vienna, 1090 Wien, AustriaNuHAG, Faculty of Mathematics, University of Vienna, 1090 Wien, AustriaThis paper introduces Gabor scattering, a feature extractor based on Gabor frames and Mallat’s scattering transform. By using a simple signal model for audio signals, specific properties of Gabor scattering are studied. It is shown that, for each layer, specific invariances to certain signal characteristics occur. Furthermore, deformation stability of the coefficient vector generated by the feature extractor is derived by using a decoupling technique which exploits the contractivity of general scattering networks. Deformations are introduced as changes in spectral shape and frequency modulation. The theoretical results are illustrated by numerical examples and experiments. Numerical evidence is given by evaluation on a synthetic and a “real” dataset, that the invariances encoded by the Gabor scattering transform lead to higher performance in comparison with just using Gabor transform, especially when few training samples are available.https://www.mdpi.com/2075-1680/8/4/106machine learningscattering transformgabor transformdeep learningtime-frequency analysiscnn |
| spellingShingle | Roswitha Bammer Monika Dörfler Pavol Harar Gabor Frames and Deep Scattering Networks in Audio Processing Axioms machine learning scattering transform gabor transform deep learning time-frequency analysis cnn |
| title | Gabor Frames and Deep Scattering Networks in Audio Processing |
| title_full | Gabor Frames and Deep Scattering Networks in Audio Processing |
| title_fullStr | Gabor Frames and Deep Scattering Networks in Audio Processing |
| title_full_unstemmed | Gabor Frames and Deep Scattering Networks in Audio Processing |
| title_short | Gabor Frames and Deep Scattering Networks in Audio Processing |
| title_sort | gabor frames and deep scattering networks in audio processing |
| topic | machine learning scattering transform gabor transform deep learning time-frequency analysis cnn |
| url | https://www.mdpi.com/2075-1680/8/4/106 |
| work_keys_str_mv | AT roswithabammer gaborframesanddeepscatteringnetworksinaudioprocessing AT monikadorfler gaborframesanddeepscatteringnetworksinaudioprocessing AT pavolharar gaborframesanddeepscatteringnetworksinaudioprocessing |