High performance speaker and vocabulary independent ASR technology for mobile phones
This paper presents the Siemens speech recognizer for mobile phones, VSR. VSR employs HMM technology and uses general-purpose phoneme-based acoustic models which make it speaker and vocabulary independent. The system can be easily reconfigured to work with arbitrary vocabularies. This provides full...
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 281 |
|---|---|
| container_issue | |
| container_start_page | II |
| container_title | |
| container_volume | 2 |
| creator | Astrov, S. Bauer, J.G. Stan, S. |
| description | This paper presents the Siemens speech recognizer for mobile phones, VSR. VSR employs HMM technology and uses general-purpose phoneme-based acoustic models which make it speaker and vocabulary independent. The system can be easily reconfigured to work with arbitrary vocabularies. This provides full flexibility for the design of the user interface which contrasts with the capabilities of other low-resource recognizers. The system requirements of VSR are very low. The emission probability calculation and the Viterbi search with a vocabulary of 30 words need only 16 MHz for real-time operation on an ARM microcontroller. The HMM acoustic models take up about 12 kilobytes of permanent storage. The most significant algorithmic improvement is the newly developed 3-D stream-based coding of the HMMs. Despite low requirements in terms of system resources VSR achieves an outstanding recognition performance. The word error rate (WER) for a recognition task with 62 German isolated words including highly confusable digits is 7.0%. |
| doi_str_mv | 10.1109/ICASSP.2003.1202349 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_1202349</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>1202349</ieee_id><sourcerecordid>1202349</sourcerecordid><originalsourceid>FETCH-ieee_primary_12023493</originalsourceid><addsrcrecordid>eNp9jsFOwzAQRFdQJELpF_SyP5CwtkMSH6sKVG6I9MCtctNtY0hsyy5I_Xty6BlpNHN4mtEALAUVQpB-eluv2va9kESqEJKkKvUNZFLVOheaPm9hoeuGJqm6qpScQSaeJeWVKPU9PKT0RURNXTYZbDf21GPgePRxNK5jTIHNN0c07oC_vjP7n8HEC1p34MCTuTOu2g88c9c7P_jTBacqjn5vB8bQe8fpEe6OZki8uOYclq8v2_Umt8y8C9GO0-Luelz9T_8APJxE6g</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>High performance speaker and vocabulary independent ASR technology for mobile phones</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Astrov, S. ; Bauer, J.G. ; Stan, S.</creator><creatorcontrib>Astrov, S. ; Bauer, J.G. ; Stan, S.</creatorcontrib><description>This paper presents the Siemens speech recognizer for mobile phones, VSR. VSR employs HMM technology and uses general-purpose phoneme-based acoustic models which make it speaker and vocabulary independent. The system can be easily reconfigured to work with arbitrary vocabularies. This provides full flexibility for the design of the user interface which contrasts with the capabilities of other low-resource recognizers. The system requirements of VSR are very low. The emission probability calculation and the Viterbi search with a vocabulary of 30 words need only 16 MHz for real-time operation on an ARM microcontroller. The HMM acoustic models take up about 12 kilobytes of permanent storage. The most significant algorithmic improvement is the newly developed 3-D stream-based coding of the HMMs. Despite low requirements in terms of system resources VSR achieves an outstanding recognition performance. The word error rate (WER) for a recognition task with 62 German isolated words including highly confusable digits is 7.0%.</description><identifier>ISSN: 1520-6149</identifier><identifier>ISBN: 9780780376632</identifier><identifier>ISBN: 0780376633</identifier><identifier>EISSN: 2379-190X</identifier><identifier>DOI: 10.1109/ICASSP.2003.1202349</identifier><language>eng</language><publisher>IEEE</publisher><subject>Acoustic emission ; Automatic speech recognition ; Hidden Markov models ; Loudspeakers ; Mobile handsets ; Probability ; Speech recognition ; User interfaces ; Viterbi algorithm ; Vocabulary</subject><ispartof>2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03), 2003, Vol.2, p.II-281</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/1202349$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2056,4047,4048,27923,54918</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/1202349$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Astrov, S.</creatorcontrib><creatorcontrib>Bauer, J.G.</creatorcontrib><creatorcontrib>Stan, S.</creatorcontrib><title>High performance speaker and vocabulary independent ASR technology for mobile phones</title><title>2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)</title><addtitle>ICASSP</addtitle><description>This paper presents the Siemens speech recognizer for mobile phones, VSR. VSR employs HMM technology and uses general-purpose phoneme-based acoustic models which make it speaker and vocabulary independent. The system can be easily reconfigured to work with arbitrary vocabularies. This provides full flexibility for the design of the user interface which contrasts with the capabilities of other low-resource recognizers. The system requirements of VSR are very low. The emission probability calculation and the Viterbi search with a vocabulary of 30 words need only 16 MHz for real-time operation on an ARM microcontroller. The HMM acoustic models take up about 12 kilobytes of permanent storage. The most significant algorithmic improvement is the newly developed 3-D stream-based coding of the HMMs. Despite low requirements in terms of system resources VSR achieves an outstanding recognition performance. The word error rate (WER) for a recognition task with 62 German isolated words including highly confusable digits is 7.0%.</description><subject>Acoustic emission</subject><subject>Automatic speech recognition</subject><subject>Hidden Markov models</subject><subject>Loudspeakers</subject><subject>Mobile handsets</subject><subject>Probability</subject><subject>Speech recognition</subject><subject>User interfaces</subject><subject>Viterbi algorithm</subject><subject>Vocabulary</subject><issn>1520-6149</issn><issn>2379-190X</issn><isbn>9780780376632</isbn><isbn>0780376633</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2003</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNp9jsFOwzAQRFdQJELpF_SyP5CwtkMSH6sKVG6I9MCtctNtY0hsyy5I_Xty6BlpNHN4mtEALAUVQpB-eluv2va9kESqEJKkKvUNZFLVOheaPm9hoeuGJqm6qpScQSaeJeWVKPU9PKT0RURNXTYZbDf21GPgePRxNK5jTIHNN0c07oC_vjP7n8HEC1p34MCTuTOu2g88c9c7P_jTBacqjn5vB8bQe8fpEe6OZki8uOYclq8v2_Umt8y8C9GO0-Luelz9T_8APJxE6g</recordid><startdate>2003</startdate><enddate>2003</enddate><creator>Astrov, S.</creator><creator>Bauer, J.G.</creator><creator>Stan, S.</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>2003</creationdate><title>High performance speaker and vocabulary independent ASR technology for mobile phones</title><author>Astrov, S. ; Bauer, J.G. ; Stan, S.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-ieee_primary_12023493</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2003</creationdate><topic>Acoustic emission</topic><topic>Automatic speech recognition</topic><topic>Hidden Markov models</topic><topic>Loudspeakers</topic><topic>Mobile handsets</topic><topic>Probability</topic><topic>Speech recognition</topic><topic>User interfaces</topic><topic>Viterbi algorithm</topic><topic>Vocabulary</topic><toplevel>online_resources</toplevel><creatorcontrib>Astrov, S.</creatorcontrib><creatorcontrib>Bauer, J.G.</creatorcontrib><creatorcontrib>Stan, S.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Astrov, S.</au><au>Bauer, J.G.</au><au>Stan, S.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>High performance speaker and vocabulary independent ASR technology for mobile phones</atitle><btitle>2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)</btitle><stitle>ICASSP</stitle><date>2003</date><risdate>2003</risdate><volume>2</volume><spage>II</spage><epage>281</epage><pages>II-281</pages><issn>1520-6149</issn><eissn>2379-190X</eissn><isbn>9780780376632</isbn><isbn>0780376633</isbn><abstract>This paper presents the Siemens speech recognizer for mobile phones, VSR. VSR employs HMM technology and uses general-purpose phoneme-based acoustic models which make it speaker and vocabulary independent. The system can be easily reconfigured to work with arbitrary vocabularies. This provides full flexibility for the design of the user interface which contrasts with the capabilities of other low-resource recognizers. The system requirements of VSR are very low. The emission probability calculation and the Viterbi search with a vocabulary of 30 words need only 16 MHz for real-time operation on an ARM microcontroller. The HMM acoustic models take up about 12 kilobytes of permanent storage. The most significant algorithmic improvement is the newly developed 3-D stream-based coding of the HMMs. Despite low requirements in terms of system resources VSR achieves an outstanding recognition performance. The word error rate (WER) for a recognition task with 62 German isolated words including highly confusable digits is 7.0%.</abstract><pub>IEEE</pub><doi>10.1109/ICASSP.2003.1202349</doi></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1520-6149 |
| ispartof | 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03), 2003, Vol.2, p.II-281 |
| issn | 1520-6149 2379-190X |
| language | eng |
| recordid | cdi_ieee_primary_1202349 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Acoustic emission Automatic speech recognition Hidden Markov models Loudspeakers Mobile handsets Probability Speech recognition User interfaces Viterbi algorithm Vocabulary |
| title | High performance speaker and vocabulary independent ASR technology for mobile phones |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-14T10%3A20%3A00IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=High%20performance%20speaker%20and%20vocabulary%20independent%20ASR%20technology%20for%20mobile%20phones&rft.btitle=2003%20IEEE%20International%20Conference%20on%20Acoustics,%20Speech,%20and%20Signal%20Processing,%202003.%20Proceedings.%20(ICASSP%20'03)&rft.au=Astrov,%20S.&rft.date=2003&rft.volume=2&rft.spage=II&rft.epage=281&rft.pages=II-281&rft.issn=1520-6149&rft.eissn=2379-190X&rft.isbn=9780780376632&rft.isbn_list=0780376633&rft_id=info:doi/10.1109/ICASSP.2003.1202349&rft_dat=%3Cieee_6IE%3E1202349%3C/ieee_6IE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=1202349&rfr_iscdi=true |