FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2183618280> ?p ?o ?g. }

• W2183618280 abstract "Automatic Speech Recognition (ASR) and Statistical Machine Translation (SMT), among other natural language processing applications, rely on a language model (LM) to provide a strong linguistic prior over word sequences of the often prohibitively large and complex hypothesis space of these systems. The language models deployed in most state-of-the-art ASR and SMT systems are n-gram models. Several statistical frameworks have been proposed to build more complex models and put (the syntactic structure of) language back into language modeling. Yet, n-gram models, despite being linguistically naive, are still favored, because estimating them from text is well understood, they are computationally efficient, and integrating them into ASR and SMT systems is straightforward. This dissertation proposes novel algorithms and techniques that make it practical to estimate and apply more complex language models in ASR and SMT tasks, in particular syntactic modes for speech recognition. While yielding significantly better performance than n-gram models, the syntactic structured language models (SLM) can not be efficiently trained on a large amount of text data due to the impractical size of the resulting model. A general information-theoretic pruning scheme is proposed to significantly reduce the size of the SLM while maintaining its prediction accuracy, therefore enabling efficient maximum likelihood estimation of the SLM parameters. The SLM, and other long-span language models, can not be directly applied during decoding or word lattice rescoring. Instead, these models are limited to an N-best rescoring framework, which as a search algorithm suffers from several known deficiencies and inefficiencies. Leveraging the theory and efficient algorithms for finite-state automata (FSA), an effective hill-climbing algorithm is developed for rescoring ASR lattices using long-span language models. It is shown that integrating the SLM into an ASR system in this manner significantly improves the WER over the computationally comparable N-best rescoring technique. Discriminative training of language models with long-span features, such as syntactic dependencies, triggers, and topic information, is limited to N-best lists for similar reasons. The FSA based hill climbing algorithm, proposed for the application of long-span models to ASR lattice rescoring, also paves the way for efficient discriminative training of long-span language models on speech lattices and thus, alleviates the shortcomings of N-best training. Long span language models—regardless of whether they are trained via maximum likelihood or trained discriminatively, and during both training and application—rely on auxiliary tools to extract nonlocal features. In the case of the SLM the syntactic features include a part of speech (POS) tagger and a parser for extracting features for each hypothesis during rescoring. These tools also slow down the application of such models in deployed systems. A general principle is presented wherein substructures common to multiple hypotheses are efficiently shared in the transition based structured prediction algorithms employed by these auxiliary tools. It is shown that the proposed substructure sharing algorithm results in substantial speedup when utilizing these tools in automatic speech recognition. The four methods and algorithms described above and detailed in this dissertation are evaluated using a state-of-the-art ASR system. It is demonstrated that these techniques could make the use of syntactically informed language models practical and hence, widespread. While this dissertation focuses on methods to efficiently apply syntactic language models to automatic speech recognition, many of the developed techniques may be used to efficiently utilize other complex language models in ASR and other applications such as machine translation." @default.
• W2183618280 created "2016-06-24" @default.
• W2183618280 creator A5008201430 @default.
• W2183618280 creator A5010122901 @default.
• W2183618280 date "2012-01-01" @default.
• W2183618280 modified "2023-09-26" @default.
• W2183618280 title "Practical and efficient incorporation of syntactic features into statistical language models" @default.
• W2183618280 cites W10704533 @default.
• W2183618280 cites W1494910745 @default.
• W2183618280 cites W1505342653 @default.
• W2183618280 cites W1508165687 @default.
• W2183618280 cites W1508977358 @default.
• W2183618280 cites W1528022942 @default.
• W2183618280 cites W1529481723 @default.
• W2183618280 cites W1543082528 @default.
• W2183618280 cites W1548538781 @default.
• W2183618280 cites W1571931074 @default.
• W2183618280 cites W1575431606 @default.
• W2183618280 cites W1582482241 @default.
• W2183618280 cites W1597533204 @default.
• W2183618280 cites W1632114991 @default.
• W2183618280 cites W1660460062 @default.
• W2183618280 cites W168280322 @default.
• W2183618280 cites W1749358154 @default.
• W2183618280 cites W179707045 @default.
• W2183618280 cites W1797288984 @default.
• W2183618280 cites W179875071 @default.
• W2183618280 cites W181643614 @default.
• W2183618280 cites W1850558006 @default.
• W2183618280 cites W1850668662 @default.
• W2183618280 cites W1859633635 @default.
• W2183618280 cites W193308610 @default.
• W2183618280 cites W1934041838 @default.
• W2183618280 cites W1966812932 @default.
• W2183618280 cites W1978671188 @default.
• W2183618280 cites W1979711143 @default.
• W2183618280 cites W1984635093 @default.
• W2183618280 cites W1989705153 @default.
• W2183618280 cites W1996903695 @default.
• W2183618280 cites W2004453227 @default.
• W2183618280 cites W2005076803 @default.
• W2183618280 cites W2007321142 @default.
• W2183618280 cites W2008652694 @default.
• W2183618280 cites W2014703436 @default.
• W2183618280 cites W2020382207 @default.
• W2183618280 cites W2024060531 @default.
• W2183618280 cites W2027979924 @default.
• W2183618280 cites W2031287088 @default.
• W2183618280 cites W2049633694 @default.
• W2183618280 cites W2050971845 @default.
• W2183618280 cites W2051434435 @default.
• W2183618280 cites W2058373514 @default.
• W2183618280 cites W2058718117 @default.
• W2183618280 cites W2075201173 @default.
• W2183618280 cites W2075601665 @default.
• W2183618280 cites W2076467305 @default.
• W2183618280 cites W2078349113 @default.
• W2183618280 cites W2081295016 @default.
• W2183618280 cites W2082092506 @default.
• W2183618280 cites W2089505529 @default.
• W2183618280 cites W2096072088 @default.
• W2183618280 cites W2098318492 @default.
• W2183618280 cites W2099111195 @default.
• W2183618280 cites W2099119623 @default.
• W2183618280 cites W2100506586 @default.
• W2183618280 cites W2104917081 @default.
• W2183618280 cites W2105852393 @default.
• W2183618280 cites W2109461733 @default.
• W2183618280 cites W2112752961 @default.
• W2183618280 cites W2113691817 @default.
• W2183618280 cites W2114858359 @default.
• W2183618280 cites W2118186095 @default.
• W2183618280 cites W2119821739 @default.
• W2183618280 cites W2124479173 @default.
• W2183618280 cites W2124772738 @default.
• W2183618280 cites W2124861326 @default.
• W2183618280 cites W2125528794 @default.
• W2183618280 cites W2126315862 @default.
• W2183618280 cites W2128201184 @default.
• W2183618280 cites W2130917146 @default.
• W2183618280 cites W2134237567 @default.
• W2183618280 cites W2134741316 @default.
• W2183618280 cites W2136504847 @default.
• W2183618280 cites W2137675115 @default.
• W2183618280 cites W2138302120 @default.
• W2183618280 cites W2140281218 @default.
• W2183618280 cites W2142416747 @default.
• W2183618280 cites W2143564602 @default.
• W2183618280 cites W2143866356 @default.
• W2183618280 cites W2148334178 @default.
• W2183618280 cites W2150051359 @default.
• W2183618280 cites W2151742940 @default.
• W2183618280 cites W2154075111 @default.
• W2183618280 cites W2155693943 @default.
• W2183618280 cites W2158025800 @default.
• W2183618280 cites W2158195707 @default.
• W2183618280 cites W2160842254 @default.
• W2183618280 cites W2167980204 @default.
• W2183618280 cites W2170986599 @default.
• W2183618280 cites W2172045870 @default.

• next