MMSE Feature Reconstruction Based on an Occlusion Model for Robust ASR View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2012

AUTHORS

José A. González , Antonio M. Peinado , Ángel M. Gómez

ABSTRACT

This paper proposes a novel compensation technique developed in the log-spectral domain. Our proposal consists in a minimum mean square error (MMSE) estimator derived from an occlusion model [1]. According to this model, the effect of noise over speech is simplified to a binary masking, so that the noise is completely masked by the speech when the speech power dominates and the other way round when the noise is dominant. As for many MMSE-based techniques, a statistical model of clean speech is required. A Gaussian mixture model is employed here. The resulting technique has clear similarities with missing-data imputation techniques although, unlike these ones, an explicit model of noise is employed by our proposal. The experimental results show the superiority of our MMSE estimator with respect to missing-data imputation with both binary and soft masks. More... »

PAGES

217-226

Book

TITLE

Advances in Speech and Language Technologies for Iberian Languages

ISBN

978-3-642-35291-1
978-3-642-35292-8

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-642-35292-8_23

DOI

http://dx.doi.org/10.1007/978-3-642-35292-8_23

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1025982675


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information and Computing Sciences", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "name": [
            "Dpt. Teor\u00eda de la Se\u00f1al, Telem\u00e1tica y Comunicaciones, Centro de Investigaci\u00f3n en Tecnolog\u00edas de la Informaci\u00f3n y de las Comunicaciones, 18071\u00a0Granada, Spain"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Gonz\u00e1lez", 
        "givenName": "Jos\u00e9 A.", 
        "id": "sg:person.010477131004.17", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010477131004.17"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "name": [
            "Dpt. Teor\u00eda de la Se\u00f1al, Telem\u00e1tica y Comunicaciones, Centro de Investigaci\u00f3n en Tecnolog\u00edas de la Informaci\u00f3n y de las Comunicaciones, 18071\u00a0Granada, Spain"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Peinado", 
        "givenName": "Antonio M.", 
        "id": "sg:person.013434670141.68", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013434670141.68"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "name": [
            "Dpt. Teor\u00eda de la Se\u00f1al, Telem\u00e1tica y Comunicaciones, Centro de Investigaci\u00f3n en Tecnolog\u00edas de la Informaci\u00f3n y de las Comunicaciones, 18071\u00a0Granada, Spain"
          ], 
          "type": "Organization"
        }, 
        "familyName": "G\u00f3mez", 
        "givenName": "\u00c1ngel M.", 
        "id": "sg:person.013366554741.94", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013366554741.94"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "https://doi.org/10.1016/j.specom.2004.03.007", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1009768526"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0167-6393(00)00034-0", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1030004932"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tasl.2007.901310", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061516022"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tsa.2003.822627", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061786206"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/asru.2005.1566472", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1093522703"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2012", 
    "datePublishedReg": "2012-01-01", 
    "description": "This paper proposes a novel compensation technique developed in the log-spectral domain. Our proposal consists in a minimum mean square error (MMSE) estimator derived from an occlusion model [1]. According to this model, the effect of noise over speech is simplified to a binary masking, so that the noise is completely masked by the speech when the speech power dominates and the other way round when the noise is dominant. As for many MMSE-based techniques, a statistical model of clean speech is required. A Gaussian mixture model is employed here. The resulting technique has clear similarities with missing-data imputation techniques although, unlike these ones, an explicit model of noise is employed by our proposal. The experimental results show the superiority of our MMSE estimator with respect to missing-data imputation with both binary and soft masks.", 
    "editor": [
      {
        "familyName": "Torre Toledano", 
        "givenName": "Doroteo", 
        "type": "Person"
      }, 
      {
        "familyName": "Ortega Gim\u00e9nez", 
        "givenName": "Alfonso", 
        "type": "Person"
      }, 
      {
        "familyName": "Teixeira", 
        "givenName": "Ant\u00f3nio", 
        "type": "Person"
      }, 
      {
        "familyName": "Gonz\u00e1lez Rodr\u00edguez", 
        "givenName": "Joaqu\u00edn", 
        "type": "Person"
      }, 
      {
        "familyName": "Hern\u00e1ndez G\u00f3mez", 
        "givenName": "Luis", 
        "type": "Person"
      }, 
      {
        "familyName": "San Segundo Hern\u00e1ndez", 
        "givenName": "Rub\u00e9n", 
        "type": "Person"
      }, 
      {
        "familyName": "Ramos Castro", 
        "givenName": "Daniel", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-642-35292-8_23", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-642-35291-1", 
        "978-3-642-35292-8"
      ], 
      "name": "Advances in Speech and Language Technologies for Iberian Languages", 
      "type": "Book"
    }, 
    "name": "MMSE Feature Reconstruction Based on an Occlusion Model for Robust ASR", 
    "pagination": "217-226", 
    "productId": [
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-642-35292-8_23"
        ]
      }, 
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "7bc8089d95b7eaa766168242a61728140ed42e50875cbc737ee7d4babe0e5a43"
        ]
      }, 
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1025982675"
        ]
      }
    ], 
    "publisher": {
      "location": "Berlin, Heidelberg", 
      "name": "Springer Berlin Heidelberg", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-642-35292-8_23", 
      "https://app.dimensions.ai/details/publication/pub.1025982675"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2019-04-15T17:13", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000001_0000000264/records_8678_00000259.jsonl", 
    "type": "Chapter", 
    "url": "http://link.springer.com/10.1007/978-3-642-35292-8_23"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-35292-8_23'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-35292-8_23'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-35292-8_23'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-35292-8_23'


 

This table displays all metadata directly associated to this object as RDF triples.

127 TRIPLES      23 PREDICATES      32 URIs      20 LITERALS      8 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-642-35292-8_23 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N701548bb071d4a9d9f0e1615034e4d0d
4 schema:citation https://doi.org/10.1016/j.specom.2004.03.007
5 https://doi.org/10.1016/s0167-6393(00)00034-0
6 https://doi.org/10.1109/asru.2005.1566472
7 https://doi.org/10.1109/tasl.2007.901310
8 https://doi.org/10.1109/tsa.2003.822627
9 schema:datePublished 2012
10 schema:datePublishedReg 2012-01-01
11 schema:description This paper proposes a novel compensation technique developed in the log-spectral domain. Our proposal consists in a minimum mean square error (MMSE) estimator derived from an occlusion model [1]. According to this model, the effect of noise over speech is simplified to a binary masking, so that the noise is completely masked by the speech when the speech power dominates and the other way round when the noise is dominant. As for many MMSE-based techniques, a statistical model of clean speech is required. A Gaussian mixture model is employed here. The resulting technique has clear similarities with missing-data imputation techniques although, unlike these ones, an explicit model of noise is employed by our proposal. The experimental results show the superiority of our MMSE estimator with respect to missing-data imputation with both binary and soft masks.
12 schema:editor N12c56f2c53054414ab3cd0b375255012
13 schema:genre chapter
14 schema:inLanguage en
15 schema:isAccessibleForFree false
16 schema:isPartOf N2257f4dca9284a4e96465afbcddaae96
17 schema:name MMSE Feature Reconstruction Based on an Occlusion Model for Robust ASR
18 schema:pagination 217-226
19 schema:productId N21e1d7d836c444cb8c8d80ce4ef069da
20 N7d56c71aa9194becb2c8290bbda7b757
21 N951499e6544a44d2b54738db6024e372
22 schema:publisher Nec91510fab74447e90f3e82e271e76f5
23 schema:sameAs https://app.dimensions.ai/details/publication/pub.1025982675
24 https://doi.org/10.1007/978-3-642-35292-8_23
25 schema:sdDatePublished 2019-04-15T17:13
26 schema:sdLicense https://scigraph.springernature.com/explorer/license/
27 schema:sdPublisher N447522fe261f4c65aaeb94ae1d05aa95
28 schema:url http://link.springer.com/10.1007/978-3-642-35292-8_23
29 sgo:license sg:explorer/license/
30 sgo:sdDataset chapters
31 rdf:type schema:Chapter
32 N054ae6a5722a4c798ed947f3de1215de schema:familyName Ortega Giménez
33 schema:givenName Alfonso
34 rdf:type schema:Person
35 N09ccaa41051947ada10f478c76d8d11c rdf:first sg:person.013434670141.68
36 rdf:rest Ne3d6368a4df8483e9a0496ce9ca7ae54
37 N12c56f2c53054414ab3cd0b375255012 rdf:first Nc1522eb71aaa407ab783890961943a06
38 rdf:rest N41e1ba88f5a74f1a89a61d65becf3b93
39 N21e1d7d836c444cb8c8d80ce4ef069da schema:name doi
40 schema:value 10.1007/978-3-642-35292-8_23
41 rdf:type schema:PropertyValue
42 N2257f4dca9284a4e96465afbcddaae96 schema:isbn 978-3-642-35291-1
43 978-3-642-35292-8
44 schema:name Advances in Speech and Language Technologies for Iberian Languages
45 rdf:type schema:Book
46 N41e1ba88f5a74f1a89a61d65becf3b93 rdf:first N054ae6a5722a4c798ed947f3de1215de
47 rdf:rest Nc941ea8c16bd4ee2a15f2ee281a7441b
48 N447522fe261f4c65aaeb94ae1d05aa95 schema:name Springer Nature - SN SciGraph project
49 rdf:type schema:Organization
50 N4b9c5fe218934dc98a0ab088d715ece4 rdf:first Nfd7bd5099f2a49188ad515f4b7bc14ac
51 rdf:rest N998f2dff216142ae9b1a7e288469cd4f
52 N567676566dcd40e3af1b4f4efb8b4bd0 rdf:first N823b33c3786b41e898ce86453688405e
53 rdf:rest rdf:nil
54 N701548bb071d4a9d9f0e1615034e4d0d rdf:first sg:person.010477131004.17
55 rdf:rest N09ccaa41051947ada10f478c76d8d11c
56 N748f194a9c8e4cfaac0d584db8499860 rdf:first Nc055a2e1c9454fd58151c78437dfe2a6
57 rdf:rest N4b9c5fe218934dc98a0ab088d715ece4
58 N7d56c71aa9194becb2c8290bbda7b757 schema:name readcube_id
59 schema:value 7bc8089d95b7eaa766168242a61728140ed42e50875cbc737ee7d4babe0e5a43
60 rdf:type schema:PropertyValue
61 N823b33c3786b41e898ce86453688405e schema:familyName Ramos Castro
62 schema:givenName Daniel
63 rdf:type schema:Person
64 N951499e6544a44d2b54738db6024e372 schema:name dimensions_id
65 schema:value pub.1025982675
66 rdf:type schema:PropertyValue
67 N998f2dff216142ae9b1a7e288469cd4f rdf:first Nbba0546932324013826e758b8881c6ac
68 rdf:rest N567676566dcd40e3af1b4f4efb8b4bd0
69 Nbba0546932324013826e758b8881c6ac schema:familyName San Segundo Hernández
70 schema:givenName Rubén
71 rdf:type schema:Person
72 Nc055a2e1c9454fd58151c78437dfe2a6 schema:familyName González Rodríguez
73 schema:givenName Joaquín
74 rdf:type schema:Person
75 Nc1522eb71aaa407ab783890961943a06 schema:familyName Torre Toledano
76 schema:givenName Doroteo
77 rdf:type schema:Person
78 Nc21a54efbdcf4e37bb8e397b97c7feb4 schema:name Dpt. Teoría de la Señal, Telemática y Comunicaciones, Centro de Investigación en Tecnologías de la Información y de las Comunicaciones, 18071 Granada, Spain
79 rdf:type schema:Organization
80 Nc941ea8c16bd4ee2a15f2ee281a7441b rdf:first Ne09d410952c541b7a973ff649c5b4ce4
81 rdf:rest N748f194a9c8e4cfaac0d584db8499860
82 Ne09d410952c541b7a973ff649c5b4ce4 schema:familyName Teixeira
83 schema:givenName António
84 rdf:type schema:Person
85 Ne3d6368a4df8483e9a0496ce9ca7ae54 rdf:first sg:person.013366554741.94
86 rdf:rest rdf:nil
87 Nec91510fab74447e90f3e82e271e76f5 schema:location Berlin, Heidelberg
88 schema:name Springer Berlin Heidelberg
89 rdf:type schema:Organisation
90 Nf1f890ec513743b08029c7aa20590177 schema:name Dpt. Teoría de la Señal, Telemática y Comunicaciones, Centro de Investigación en Tecnologías de la Información y de las Comunicaciones, 18071 Granada, Spain
91 rdf:type schema:Organization
92 Nf4261f912ca643f0bb7ecd4384efcde0 schema:name Dpt. Teoría de la Señal, Telemática y Comunicaciones, Centro de Investigación en Tecnologías de la Información y de las Comunicaciones, 18071 Granada, Spain
93 rdf:type schema:Organization
94 Nfd7bd5099f2a49188ad515f4b7bc14ac schema:familyName Hernández Gómez
95 schema:givenName Luis
96 rdf:type schema:Person
97 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
98 schema:name Information and Computing Sciences
99 rdf:type schema:DefinedTerm
100 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
101 schema:name Artificial Intelligence and Image Processing
102 rdf:type schema:DefinedTerm
103 sg:person.010477131004.17 schema:affiliation Nc21a54efbdcf4e37bb8e397b97c7feb4
104 schema:familyName González
105 schema:givenName José A.
106 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010477131004.17
107 rdf:type schema:Person
108 sg:person.013366554741.94 schema:affiliation Nf4261f912ca643f0bb7ecd4384efcde0
109 schema:familyName Gómez
110 schema:givenName Ángel M.
111 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013366554741.94
112 rdf:type schema:Person
113 sg:person.013434670141.68 schema:affiliation Nf1f890ec513743b08029c7aa20590177
114 schema:familyName Peinado
115 schema:givenName Antonio M.
116 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013434670141.68
117 rdf:type schema:Person
118 https://doi.org/10.1016/j.specom.2004.03.007 schema:sameAs https://app.dimensions.ai/details/publication/pub.1009768526
119 rdf:type schema:CreativeWork
120 https://doi.org/10.1016/s0167-6393(00)00034-0 schema:sameAs https://app.dimensions.ai/details/publication/pub.1030004932
121 rdf:type schema:CreativeWork
122 https://doi.org/10.1109/asru.2005.1566472 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093522703
123 rdf:type schema:CreativeWork
124 https://doi.org/10.1109/tasl.2007.901310 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061516022
125 rdf:type schema:CreativeWork
126 https://doi.org/10.1109/tsa.2003.822627 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061786206
127 rdf:type schema:CreativeWork
 




Preview window. Press ESC to close (or click here)


...