dorsal/arxiv
View SchemaDelay-Coordinates Embeddings as a Data Mining Tool for Denoising Speech Signals
| Authors | D. Napoletani, C. A. Berenstein, T. Sauer, D. C. Struppa, D. Walnut |
|---|---|
| Categories | |
| ArXiv ID | physics/0504155 |
| URL | https://arxiv.org/abs/physics/0504155 |
| DOI | 10.1063/1.2384909 |
| Journal | Chaos 16, 043116 (2006) |
Abstract
In this paper we utilize techniques from the theory of non-linear dynamical systems to define a notion of embedding threshold estimators. More specifically we use delay-coordinates embeddings of sets of coefficients of the measured signal (in some chosen frame) as a data mining tool to separate structures that are likely to be generated by signals belonging to some predetermined data set. We describe a particular variation of the embedding threshold estimator implemented in a windowed Fourier frame, and we apply it to speech signals heavily corrupted with the addition of several types of white noise. Our experimental work seems to suggest that, after training on the data sets of interest,these estimators perform well for a variety of white noise processes and noise intensity levels. The method is compared, for the case of Gaussian white noise, to a block thresholding estimator.
{
"annotation_id": "d8e2ace8-dd71-4876-ac8d-803456c39b8d",
"date_created": "2026-03-02T18:00:57.171000Z",
"date_modified": "2026-03-02T18:00:57.171000Z",
"file_hash": "b3b024cb5ee296bbda8e234530c8084c39710a608d5987a7beaf678718d6da89",
"private": false,
"record": {
"abstract": "In this paper we utilize techniques from the theory of non-linear dynamical\nsystems to define a notion of embedding threshold estimators. More specifically\nwe use delay-coordinates embeddings of sets of coefficients of the measured\nsignal (in some chosen frame) as a data mining tool to separate structures that\nare likely to be generated by signals belonging to some predetermined data set.\nWe describe a particular variation of the embedding threshold estimator\nimplemented in a windowed Fourier frame, and we apply it to speech signals\nheavily corrupted with the addition of several types of white noise. Our\nexperimental work seems to suggest that, after training on the data sets of\ninterest,these estimators perform well for a variety of white noise processes\nand noise intensity levels. The method is compared, for the case of Gaussian\nwhite noise, to a block thresholding estimator.",
"arxiv_id": "physics/0504155",
"authors": [
"D. Napoletani",
"C. A. Berenstein",
"T. Sauer",
"D. C. Struppa",
"D. Walnut"
],
"categories": [
"physics.data-an",
"nlin.CD"
],
"doi": "10.1063/1.2384909",
"journal_ref": "Chaos 16, 043116 (2006)",
"title": "Delay-Coordinates Embeddings as a Data Mining Tool for Denoising Speech Signals",
"url": "https://arxiv.org/abs/physics/0504155"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "34c66fff-2ce8-417e-9a57-85967583ca68",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}