{"id":"https://openalex.org/W2295078202","doi":"https://doi.org/10.21437/interspeech.2014-165","title":"Comparing approaches to convert recurrent neural networks into backoff language models for efficient decoding","display_name":"Comparing approaches to convert recurrent neural networks into backoff language models for efficient decoding","publication_year":2014,"publication_date":"2014-09-14","ids":{"openalex":"https://openalex.org/W2295078202","doi":"https://doi.org/10.21437/interspeech.2014-165","mag":"2295078202"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2014-165","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2014-165","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2014","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5055589412","display_name":"Heike Adel","orcid":"https://orcid.org/0000-0003-2787-0084"},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Heike Adel","raw_affiliation_strings":["Karlsruhe Institute of Technology"],"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050325468","display_name":"Katrin Kirchhoff","orcid":"https://orcid.org/0000-0002-6645-6030"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Katrin Kirchhoff","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020700841","display_name":"Ngoc Thang Vu","orcid":"https://orcid.org/0000-0001-7893-9147"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ngoc Thang Vu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087799432","display_name":"Dominic Telaar","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dominic Telaar","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5058049725","display_name":"Tanja Schultz","orcid":"https://orcid.org/0000-0002-9809-7028"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tanja Schultz","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5055589412"],"corresponding_institution_ids":["https://openalex.org/I102335020"],"apc_list":null,"apc_paid":null,"fwci":2.4541,"has_fulltext":false,"cited_by_count":24,"citation_normalized_percentile":{"value":0.91323711,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"651","last_page":"655"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9976999759674072,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9962999820709229,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/perplexity","display_name":"Perplexity","score":0.9843836426734924},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8530078530311584},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.7567887306213379},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.7106949687004089},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5763033032417297},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5126916170120239},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.5067370533943176},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.471938818693161},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.44304758310317993},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4301658272743225},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.42406165599823},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.3235211968421936},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3216337561607361},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.23817086219787598},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.07824292778968811}],"concepts":[{"id":"https://openalex.org/C100279451","wikidata":"https://www.wikidata.org/wiki/Q372193","display_name":"Perplexity","level":3,"score":0.9843836426734924},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8530078530311584},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.7567887306213379},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.7106949687004089},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5763033032417297},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5126916170120239},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.5067370533943176},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.471938818693161},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.44304758310317993},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4301658272743225},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.42406165599823},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3235211968421936},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3216337561607361},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.23817086219787598},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.07824292778968811},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2014-165","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2014-165","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2014","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.6299999952316284,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W118155631","https://openalex.org/W179875071","https://openalex.org/W197277712","https://openalex.org/W1604771987","https://openalex.org/W1631260214","https://openalex.org/W1979625042","https://openalex.org/W2098355803","https://openalex.org/W2110415041","https://openalex.org/W2153433699","https://openalex.org/W2158195707","https://openalex.org/W2171928131","https://openalex.org/W2405047074","https://openalex.org/W2407682997","https://openalex.org/W4285719527"],"related_works":["https://openalex.org/W2610308580","https://openalex.org/W2169518243","https://openalex.org/W2252095989","https://openalex.org/W4322096525","https://openalex.org/W2551914602","https://openalex.org/W4287323699","https://openalex.org/W4281893144","https://openalex.org/W2050138804","https://openalex.org/W2151348424","https://openalex.org/W1494910745"],"abstract_inverted_index":{"In":[0,149],"this":[1],"paper,":[2],"we":[3],"investigate":[4],"and":[5,86,100,163],"compare":[6],"three":[7,75,125],"different":[8,76],"possibilities":[9],"to":[10,57,64,134],"convert":[11,58],"recurrent":[12],"neural":[13],"network":[14],"language":[15,20,34,92],"models":[16,21,93],"(RNNLMs)":[17],"into":[18,60,68],"backoff":[19],"(BNLM).":[22],"While":[23],"RNNLMs":[24],"often":[25],"outperform":[26],"traditional":[27],"n-gram":[28],"approaches":[29],"in":[30,48,62,96,104],"the":[31,69,105,108,114,121,127,138,145,151,160,164,172],"task":[32],"of":[33,55,98,107,123],"modeling,":[35],"their":[36,66],"computational":[37,147],"demands":[38],"make":[39],"them":[40,59],"unsuitable":[41],"for":[42],"an":[43,49,87],"efficient":[44],"usage":[45],"during":[46],"decoding":[47,70],"LVCSR":[50],"system.":[51],"It":[52],"is,":[53],"therefore,":[54],"interest":[56],"BNLMs":[61],"order":[63],"integrate":[65],"information":[67],"process.":[71],"This":[72],"paper":[73],"compares":[74],"approaches:":[77],"a":[78,82],"text":[79,128],"based":[80,84,129],"conversion,":[81],"probability":[83],"conversion":[85,130],"iterative":[88],"conversion.":[89],"The":[90],"resulting":[91],"are":[94,117],"evaluated":[95],"terms":[97],"perplexity":[99,152],"mixed":[101,165],"error":[102,166],"rate":[103,167],"context":[106],"Code-Switching":[109],"data":[110,174],"corpus":[111,140],"SEAME.":[112],"Although":[113],"best":[115],"results":[116,122],"obtained":[118],"by":[119,156,168],"combining":[120],"all":[124],"approaches,":[126],"approach":[131],"alone":[132],"leads":[133],"significant":[135],"improvements":[136],"on":[137,159,171],"SEAME":[139],"as":[141],"well":[142],"while":[143],"offering":[144],"highest":[146],"efficiency.":[148],"total,":[150],"can":[153],"be":[154],"reduced":[155],"11.4%":[157],"relative":[158,170],"evaluation":[161],"set":[162],"3.0%":[169],"same":[173],"set.":[175]},"counts_by_year":[{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":1},{"year":2020,"cited_by_count":3},{"year":2019,"cited_by_count":7},{"year":2018,"cited_by_count":5},{"year":2017,"cited_by_count":2},{"year":2016,"cited_by_count":1},{"year":2014,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}