Acoustic Modelling Using Continuous Rational Kernels
Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors....
Ausführliche Beschreibung
Gespeichert in:
| Autor*in: |
Layton, Martin [verfasserIn] Gales, Mark [verfasserIn] |
|---|
| Format: |
E-Artikel |
|---|---|
| Sprache: |
Englisch |
| Erschienen: |
2007 |
|---|
| Schlagwörter: |
|---|
| Übergeordnetes Werk: |
Enthalten in: Journal of VLSI signal processing systems for signal, image and video technology - Springer Netherlands, 1989, 48(2007), 1-2 vom: 05. Mai, Seite 67-82 |
|---|---|
| Übergeordnetes Werk: |
volume:48 ; year:2007 ; number:1-2 ; day:05 ; month:05 ; pages:67-82 |
| Links: |
|---|
| DOI / URN: |
10.1007/s11265-006-0027-4 |
|---|
| Katalog-ID: |
SPR018319009 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | SPR018319009 | ||
| 003 | DE-627 | ||
| 005 | 20201124222358.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 201006s2007 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1007/s11265-006-0027-4 |2 doi | |
| 035 | |a (DE-627)SPR018319009 | ||
| 035 | |a (SPR)s11265-006-0027-4-e | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Layton, Martin |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Acoustic Modelling Using Continuous Rational Kernels |
| 264 | 1 | |c 2007 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 520 | |a Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented. | ||
| 650 | 4 | |a augmented statistical models |7 (dpeaa)DE-He213 | |
| 650 | 4 | |a rational kernels |7 (dpeaa)DE-He213 | |
| 650 | 4 | |a speech recognition |7 (dpeaa)DE-He213 | |
| 650 | 4 | |a TIMIT database |7 (dpeaa)DE-He213 | |
| 700 | 1 | |a Gales, Mark |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Journal of VLSI signal processing systems for signal, image and video technology |d Springer Netherlands, 1989 |g 48(2007), 1-2 vom: 05. Mai, Seite 67-82 |w (DE-627)SPR018308090 |7 nnns |
| 773 | 1 | 8 | |g volume:48 |g year:2007 |g number:1-2 |g day:05 |g month:05 |g pages:67-82 |
| 856 | 4 | 0 | |u https://dx.doi.org/10.1007/s11265-006-0027-4 |z lizenzpflichtig |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a SYSFLAG_A | ||
| 912 | |a GBV_SPRINGER | ||
| 912 | |a GBV_ILN_40 | ||
| 912 | |a GBV_ILN_2006 | ||
| 912 | |a GBV_ILN_2027 | ||
| 951 | |a AR | ||
| 952 | |d 48 |j 2007 |e 1-2 |b 05 |c 05 |h 67-82 | ||
| author_variant |
m l ml m g mg |
|---|---|
| matchkey_str |
laytonmartingalesmark:2007----:cutcoelnuigotnos |
| hierarchy_sort_str |
2007 |
| publishDate |
2007 |
| allfields |
10.1007/s11265-006-0027-4 doi (DE-627)SPR018319009 (SPR)s11265-006-0027-4-e DE-627 ger DE-627 rakwb eng Layton, Martin verfasserin aut Acoustic Modelling Using Continuous Rational Kernels 2007 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented. augmented statistical models (dpeaa)DE-He213 rational kernels (dpeaa)DE-He213 speech recognition (dpeaa)DE-He213 TIMIT database (dpeaa)DE-He213 Gales, Mark verfasserin aut Enthalten in Journal of VLSI signal processing systems for signal, image and video technology Springer Netherlands, 1989 48(2007), 1-2 vom: 05. Mai, Seite 67-82 (DE-627)SPR018308090 nnns volume:48 year:2007 number:1-2 day:05 month:05 pages:67-82 https://dx.doi.org/10.1007/s11265-006-0027-4 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_SPRINGER GBV_ILN_40 GBV_ILN_2006 GBV_ILN_2027 AR 48 2007 1-2 05 05 67-82 |
| spelling |
10.1007/s11265-006-0027-4 doi (DE-627)SPR018319009 (SPR)s11265-006-0027-4-e DE-627 ger DE-627 rakwb eng Layton, Martin verfasserin aut Acoustic Modelling Using Continuous Rational Kernels 2007 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented. augmented statistical models (dpeaa)DE-He213 rational kernels (dpeaa)DE-He213 speech recognition (dpeaa)DE-He213 TIMIT database (dpeaa)DE-He213 Gales, Mark verfasserin aut Enthalten in Journal of VLSI signal processing systems for signal, image and video technology Springer Netherlands, 1989 48(2007), 1-2 vom: 05. Mai, Seite 67-82 (DE-627)SPR018308090 nnns volume:48 year:2007 number:1-2 day:05 month:05 pages:67-82 https://dx.doi.org/10.1007/s11265-006-0027-4 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_SPRINGER GBV_ILN_40 GBV_ILN_2006 GBV_ILN_2027 AR 48 2007 1-2 05 05 67-82 |
| allfields_unstemmed |
10.1007/s11265-006-0027-4 doi (DE-627)SPR018319009 (SPR)s11265-006-0027-4-e DE-627 ger DE-627 rakwb eng Layton, Martin verfasserin aut Acoustic Modelling Using Continuous Rational Kernels 2007 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented. augmented statistical models (dpeaa)DE-He213 rational kernels (dpeaa)DE-He213 speech recognition (dpeaa)DE-He213 TIMIT database (dpeaa)DE-He213 Gales, Mark verfasserin aut Enthalten in Journal of VLSI signal processing systems for signal, image and video technology Springer Netherlands, 1989 48(2007), 1-2 vom: 05. Mai, Seite 67-82 (DE-627)SPR018308090 nnns volume:48 year:2007 number:1-2 day:05 month:05 pages:67-82 https://dx.doi.org/10.1007/s11265-006-0027-4 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_SPRINGER GBV_ILN_40 GBV_ILN_2006 GBV_ILN_2027 AR 48 2007 1-2 05 05 67-82 |
| allfieldsGer |
10.1007/s11265-006-0027-4 doi (DE-627)SPR018319009 (SPR)s11265-006-0027-4-e DE-627 ger DE-627 rakwb eng Layton, Martin verfasserin aut Acoustic Modelling Using Continuous Rational Kernels 2007 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented. augmented statistical models (dpeaa)DE-He213 rational kernels (dpeaa)DE-He213 speech recognition (dpeaa)DE-He213 TIMIT database (dpeaa)DE-He213 Gales, Mark verfasserin aut Enthalten in Journal of VLSI signal processing systems for signal, image and video technology Springer Netherlands, 1989 48(2007), 1-2 vom: 05. Mai, Seite 67-82 (DE-627)SPR018308090 nnns volume:48 year:2007 number:1-2 day:05 month:05 pages:67-82 https://dx.doi.org/10.1007/s11265-006-0027-4 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_SPRINGER GBV_ILN_40 GBV_ILN_2006 GBV_ILN_2027 AR 48 2007 1-2 05 05 67-82 |
| allfieldsSound |
10.1007/s11265-006-0027-4 doi (DE-627)SPR018319009 (SPR)s11265-006-0027-4-e DE-627 ger DE-627 rakwb eng Layton, Martin verfasserin aut Acoustic Modelling Using Continuous Rational Kernels 2007 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented. augmented statistical models (dpeaa)DE-He213 rational kernels (dpeaa)DE-He213 speech recognition (dpeaa)DE-He213 TIMIT database (dpeaa)DE-He213 Gales, Mark verfasserin aut Enthalten in Journal of VLSI signal processing systems for signal, image and video technology Springer Netherlands, 1989 48(2007), 1-2 vom: 05. Mai, Seite 67-82 (DE-627)SPR018308090 nnns volume:48 year:2007 number:1-2 day:05 month:05 pages:67-82 https://dx.doi.org/10.1007/s11265-006-0027-4 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_SPRINGER GBV_ILN_40 GBV_ILN_2006 GBV_ILN_2027 AR 48 2007 1-2 05 05 67-82 |
| language |
English |
| source |
Enthalten in Journal of VLSI signal processing systems for signal, image and video technology 48(2007), 1-2 vom: 05. Mai, Seite 67-82 volume:48 year:2007 number:1-2 day:05 month:05 pages:67-82 |
| sourceStr |
Enthalten in Journal of VLSI signal processing systems for signal, image and video technology 48(2007), 1-2 vom: 05. Mai, Seite 67-82 volume:48 year:2007 number:1-2 day:05 month:05 pages:67-82 |
| format_phy_str_mv |
Article |
| institution |
findex.gbv.de |
| topic_facet |
augmented statistical models rational kernels speech recognition TIMIT database |
| isfreeaccess_bool |
false |
| container_title |
Journal of VLSI signal processing systems for signal, image and video technology |
| authorswithroles_txt_mv |
Layton, Martin @@aut@@ Gales, Mark @@aut@@ |
| publishDateDaySort_date |
2007-05-05T00:00:00Z |
| hierarchy_top_id |
SPR018308090 |
| id |
SPR018319009 |
| language_de |
englisch |
| fullrecord |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">SPR018319009</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20201124222358.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">201006s2007 xx |||||o 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1007/s11265-006-0027-4</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)SPR018319009</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(SPR)s11265-006-0027-4-e</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Layton, Martin</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Acoustic Modelling Using Continuous Rational Kernels</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2007</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">Computermedien</subfield><subfield code="b">c</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield><subfield code="b">cr</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">augmented statistical models</subfield><subfield code="7">(dpeaa)DE-He213</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">rational kernels</subfield><subfield code="7">(dpeaa)DE-He213</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">speech recognition</subfield><subfield code="7">(dpeaa)DE-He213</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">TIMIT database</subfield><subfield code="7">(dpeaa)DE-He213</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Gales, Mark</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Journal of VLSI signal processing systems for signal, image and video technology</subfield><subfield code="d">Springer Netherlands, 1989</subfield><subfield code="g">48(2007), 1-2 vom: 05. Mai, Seite 67-82</subfield><subfield code="w">(DE-627)SPR018308090</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:48</subfield><subfield code="g">year:2007</subfield><subfield code="g">number:1-2</subfield><subfield code="g">day:05</subfield><subfield code="g">month:05</subfield><subfield code="g">pages:67-82</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://dx.doi.org/10.1007/s11265-006-0027-4</subfield><subfield code="z">lizenzpflichtig</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_SPRINGER</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_40</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2006</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2027</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">48</subfield><subfield code="j">2007</subfield><subfield code="e">1-2</subfield><subfield code="b">05</subfield><subfield code="c">05</subfield><subfield code="h">67-82</subfield></datafield></record></collection>
|
| author |
Layton, Martin |
| spellingShingle |
Layton, Martin misc augmented statistical models misc rational kernels misc speech recognition misc TIMIT database Acoustic Modelling Using Continuous Rational Kernels |
| authorStr |
Layton, Martin |
| ppnlink_with_tag_str_mv |
@@773@@(DE-627)SPR018308090 |
| format |
electronic Article |
| delete_txt_mv |
keep |
| author_role |
aut aut |
| collection |
springer |
| remote_str |
true |
| illustrated |
Not Illustrated |
| topic_title |
Acoustic Modelling Using Continuous Rational Kernels augmented statistical models (dpeaa)DE-He213 rational kernels (dpeaa)DE-He213 speech recognition (dpeaa)DE-He213 TIMIT database (dpeaa)DE-He213 |
| topic |
misc augmented statistical models misc rational kernels misc speech recognition misc TIMIT database |
| topic_unstemmed |
misc augmented statistical models misc rational kernels misc speech recognition misc TIMIT database |
| topic_browse |
misc augmented statistical models misc rational kernels misc speech recognition misc TIMIT database |
| format_facet |
Elektronische Aufsätze Aufsätze Elektronische Ressource |
| format_main_str_mv |
Text Zeitschrift/Artikel |
| carriertype_str_mv |
cr |
| hierarchy_parent_title |
Journal of VLSI signal processing systems for signal, image and video technology |
| hierarchy_parent_id |
SPR018308090 |
| hierarchy_top_title |
Journal of VLSI signal processing systems for signal, image and video technology |
| isfreeaccess_txt |
false |
| familylinks_str_mv |
(DE-627)SPR018308090 |
| title |
Acoustic Modelling Using Continuous Rational Kernels |
| ctrlnum |
(DE-627)SPR018319009 (SPR)s11265-006-0027-4-e |
| title_full |
Acoustic Modelling Using Continuous Rational Kernels |
| author_sort |
Layton, Martin |
| journal |
Journal of VLSI signal processing systems for signal, image and video technology |
| journalStr |
Journal of VLSI signal processing systems for signal, image and video technology |
| lang_code |
eng |
| isOA_bool |
false |
| recordtype |
marc |
| publishDateSort |
2007 |
| contenttype_str_mv |
txt |
| container_start_page |
67 |
| author_browse |
Layton, Martin Gales, Mark |
| container_volume |
48 |
| format_se |
Elektronische Aufsätze |
| author-letter |
Layton, Martin |
| doi_str_mv |
10.1007/s11265-006-0027-4 |
| author2-role |
verfasserin |
| title_sort |
acoustic modelling using continuous rational kernels |
| title_auth |
Acoustic Modelling Using Continuous Rational Kernels |
| abstract |
Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented. |
| abstractGer |
Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented. |
| abstract_unstemmed |
Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented. |
| collection_details |
GBV_USEFLAG_A SYSFLAG_A GBV_SPRINGER GBV_ILN_40 GBV_ILN_2006 GBV_ILN_2027 |
| container_issue |
1-2 |
| title_short |
Acoustic Modelling Using Continuous Rational Kernels |
| url |
https://dx.doi.org/10.1007/s11265-006-0027-4 |
| remote_bool |
true |
| author2 |
Gales, Mark |
| author2Str |
Gales, Mark |
| ppnlink |
SPR018308090 |
| mediatype_str_mv |
c |
| isOA_txt |
false |
| hochschulschrift_bool |
false |
| doi_str |
10.1007/s11265-006-0027-4 |
| up_date |
2024-07-03T18:53:11.407Z |
| _version_ |
1803585104287629312 |
| fullrecord_marcxml |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">SPR018319009</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20201124222358.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">201006s2007 xx |||||o 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1007/s11265-006-0027-4</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)SPR018319009</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(SPR)s11265-006-0027-4-e</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Layton, Martin</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Acoustic Modelling Using Continuous Rational Kernels</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2007</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">Computermedien</subfield><subfield code="b">c</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield><subfield code="b">cr</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract Many discriminative classification algorithms are designed for tasks where samples can be represented by fixed-length vectors. However, many examples in the fields of text processing, computational biology and speech recognition are best represented as variable-length sequences of vectors. Although several dynamic kernels have been proposed for mapping sequences of discrete observations into fixed-dimensional feature-spaces, few kernels exist for sequences of continuous observations. This paper introduces continuous rational kernels, an extension of standard rational kernels, as a general framework for classifying sequences of continuous observations. In addition to allowing new task-dependent kernels to be defined, continuous rational kernels allow existing continuous dynamic kernels, such as Fisher and generative kernels, to be calculated using standard weighted finite-state transducer algorithms. Preliminary results on both a large vocabulary continuous speech recognition (LVCSR) task and the TIMIT database are presented.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">augmented statistical models</subfield><subfield code="7">(dpeaa)DE-He213</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">rational kernels</subfield><subfield code="7">(dpeaa)DE-He213</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">speech recognition</subfield><subfield code="7">(dpeaa)DE-He213</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">TIMIT database</subfield><subfield code="7">(dpeaa)DE-He213</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Gales, Mark</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Journal of VLSI signal processing systems for signal, image and video technology</subfield><subfield code="d">Springer Netherlands, 1989</subfield><subfield code="g">48(2007), 1-2 vom: 05. Mai, Seite 67-82</subfield><subfield code="w">(DE-627)SPR018308090</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:48</subfield><subfield code="g">year:2007</subfield><subfield code="g">number:1-2</subfield><subfield code="g">day:05</subfield><subfield code="g">month:05</subfield><subfield code="g">pages:67-82</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://dx.doi.org/10.1007/s11265-006-0027-4</subfield><subfield code="z">lizenzpflichtig</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_SPRINGER</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_40</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2006</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2027</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">48</subfield><subfield code="j">2007</subfield><subfield code="e">1-2</subfield><subfield code="b">05</subfield><subfield code="c">05</subfield><subfield code="h">67-82</subfield></datafield></record></collection>
|
| score |
7.3995867 |