Matches in SemOpenAlex for { <https://semopenalex.org/work/W7082836> ?p ?o ?g. }
- W7082836 abstract "This dissertation considers the problem of information retrieval in speech. Today's speech retrieval systems generally use a large vocabulary continuous speech recognition system to first hypothesize the words which were spoken. Because these systems have a predefined lexicon, words which fall outside of the lexicon can significantly reduce search quality|as measured by Mean Average Precision (MAP). This is particularly important because these Out-Of-Vocabulary (OOV) words are often rare and therefore good discriminators for topically relevant speech segments. The focus of this dissertation is on handling these out-of-vocabulary query words. The approach is to combine results from a word-based speech retrieval system with those from vocabulary-independent ranked utterance retrieval. The goal of ranked utterance retrieval is to rank speech utterances by the system's confidence that they contain a particular spoken word, which is accomplished by ranking the utterances by the estimated frequency of the word in the utterance. Several new approaches for estimating this frequency are considered, which are motivated by the disparity between reference and errorfully hypothesized phoneme sequences. The first method learns alternate pronunciations or degradations from actual recognition hypotheses and incorporates these variants into a new generative estimator for term frequency. A second method learns transformations of several easily computed features in a discriminative model for the same task. Both methods significantly improved ranked utterance retrieval in an experimental validation on new speech. The best of these ranked utterance retrieval methods is then combined with a word-based speech retrieval system. The combination approach uses a normalization learned in an additive model, which maps the retrieval status values from each system into estimated probabilities of relevance that are easily combined. Using this combination, much of the MAP lost because of OOV words is recovered. Evaluated on a collection of spontaneous, conversational speech, the system recovers 57.5% of the MAP lost on short (title-only) queries and 41.3% on longer (title plus description) queries." @default.
- W7082836 created "2016-06-24" @default.
- W7082836 creator A5074322406 @default.
- W7082836 creator A5077897564 @default.
- W7082836 date "2008-01-01" @default.
- W7082836 modified "2023-09-26" @default.
- W7082836 title "Combining evidence from unconstrained spoken term frequency estimation for improved speech retrieval" @default.
- W7082836 cites W115020434 @default.
- W7082836 cites W1482214997 @default.
- W7082836 cites W1513618424 @default.
- W7082836 cites W1540841176 @default.
- W7082836 cites W1556631438 @default.
- W7082836 cites W160722039 @default.
- W7082836 cites W1631260214 @default.
- W7082836 cites W182831726 @default.
- W7082836 cites W183136171 @default.
- W7082836 cites W1880014881 @default.
- W7082836 cites W19563386 @default.
- W7082836 cites W1978604688 @default.
- W7082836 cites W2020237802 @default.
- W7082836 cites W2028282769 @default.
- W7082836 cites W2037213523 @default.
- W7082836 cites W2037320173 @default.
- W7082836 cites W2038045014 @default.
- W7082836 cites W2048045485 @default.
- W7082836 cites W2049633694 @default.
- W7082836 cites W2050309898 @default.
- W7082836 cites W2060494110 @default.
- W7082836 cites W2076471919 @default.
- W7082836 cites W2078396654 @default.
- W7082836 cites W2080141381 @default.
- W7082836 cites W2086253379 @default.
- W7082836 cites W2097670688 @default.
- W7082836 cites W2104765974 @default.
- W7082836 cites W2105079665 @default.
- W7082836 cites W2110758086 @default.
- W7082836 cites W2113641473 @default.
- W7082836 cites W2113788796 @default.
- W7082836 cites W2115400587 @default.
- W7082836 cites W2116394790 @default.
- W7082836 cites W2120708938 @default.
- W7082836 cites W2124807415 @default.
- W7082836 cites W2125838338 @default.
- W7082836 cites W2127923419 @default.
- W7082836 cites W2138370049 @default.
- W7082836 cites W2139839651 @default.
- W7082836 cites W2140308441 @default.
- W7082836 cites W2143369964 @default.
- W7082836 cites W2145790509 @default.
- W7082836 cites W2146081744 @default.
- W7082836 cites W2146815834 @default.
- W7082836 cites W2153653739 @default.
- W7082836 cites W2155427043 @default.
- W7082836 cites W2168330663 @default.
- W7082836 cites W2170907675 @default.
- W7082836 cites W2175238961 @default.
- W7082836 cites W2186490579 @default.
- W7082836 cites W2403397815 @default.
- W7082836 cites W2797583072 @default.
- W7082836 cites W2915143788 @default.
- W7082836 cites W2917989997 @default.
- W7082836 cites W2950186769 @default.
- W7082836 cites W49437105 @default.
- W7082836 cites W56827455 @default.
- W7082836 cites W82149049 @default.
- W7082836 cites W85622666 @default.
- W7082836 cites W2075294717 @default.
- W7082836 cites W2520132380 @default.
- W7082836 hasPublicationYear "2008" @default.
- W7082836 type Work @default.
- W7082836 sameAs 7082836 @default.
- W7082836 citedByCount "1" @default.
- W7082836 crossrefType "dissertation" @default.
- W7082836 hasAuthorship W7082836A5074322406 @default.
- W7082836 hasAuthorship W7082836A5077897564 @default.
- W7082836 hasConcept C121332964 @default.
- W7082836 hasConcept C138885662 @default.
- W7082836 hasConcept C14999030 @default.
- W7082836 hasConcept C154945302 @default.
- W7082836 hasConcept C189430467 @default.
- W7082836 hasConcept C204321447 @default.
- W7082836 hasConcept C2775852435 @default.
- W7082836 hasConcept C2777601683 @default.
- W7082836 hasConcept C2778121359 @default.
- W7082836 hasConcept C28490314 @default.
- W7082836 hasConcept C41008148 @default.
- W7082836 hasConcept C41895202 @default.
- W7082836 hasConcept C61797465 @default.
- W7082836 hasConcept C62520636 @default.
- W7082836 hasConcept C90805587 @default.
- W7082836 hasConcept C91863865 @default.
- W7082836 hasConcept C97931131 @default.
- W7082836 hasConceptScore W7082836C121332964 @default.
- W7082836 hasConceptScore W7082836C138885662 @default.
- W7082836 hasConceptScore W7082836C14999030 @default.
- W7082836 hasConceptScore W7082836C154945302 @default.
- W7082836 hasConceptScore W7082836C189430467 @default.
- W7082836 hasConceptScore W7082836C204321447 @default.
- W7082836 hasConceptScore W7082836C2775852435 @default.
- W7082836 hasConceptScore W7082836C2777601683 @default.