Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 26 for “"Subwords"”.

  1. The use of subwords for Automatic Speech Recognition

    Talgreining er notuð til að umrita töluðu máli í texta með reiknilíkönum. Hefðbundnir talgreinar notast við heil orð sem eininguna sem kerfið á skila en það gerir það að verkum að kerfin verða takmörkuð af þeim orðum sem líkaninu er gefið við þjálfun. Tungumál eru lifandi, ný orð eru sífellt að …

    reykjavik Repository record for The use of subwords for Automatic Speech Recognition (opens in a new tab)

  2. Words and Subwords: Phonology in a Piece-Based Syntactic Morphology

    The goal of this dissertation is to take generalizations made in a variety of phonological and morphological theories and account for them in a piece-based syntactic theory of morphology. The theories discussed are Cyclic phonology, Lexical Phonology (and Stratal Optimality Theory), Prosodic …

    penn Repository record for Words and Subwords: Phonology in a Piece-Based Syntactic Morphology (opens in a new tab)

  3. Text mining at multiple granularity: leveraging subwords, words, phrases, and sentences

    … granularity. Operating on these fine-grained subwords addresses many important problems in NLP namely the long-tail data-sparsity problem whereby most words in a corpus are infrequent, and the more severe out-of-vocabulary problem. To effectively and efficiently mine these subword structures, …

    uiuc Repository record for Text mining at multiple granularity: leveraging subwords, words, phrases, and sentences (opens in a new tab)

  4. A Generalization of Sturmian Sequences: Combinatorial Structure and Transcendence

    … Ak = {1,2,...,k} having (k - 1)n + 1 distinct subwords of length n. These sequences, originally defined by P. Arnoux and G. Rauzy, are a natural generalization of binary Sturmian sequences. We describe two simple combinatorial algorithms for constructing characteristic Arnoux-Rauzy sequences …

    unt Repository record for A Generalization of Sturmian Sequences: Combinatorial Structure and Transcendence (opens in a new tab)

  5. Limits of Rauzy Graphs of Low-Complexity Words

    … a finite alphabet), for which the number of subwords of length n is bounded by some Kn --- examples of such a word include the Thue-Morse word 01101001... and the Fibonacci word. Rauzy graphs Rn (omega) have the length n subwords of omega as vertices, and the oriented edges between vertices …

    ottawa-retro Repository record for Limits of Rauzy Graphs of Low-Complexity Words (opens in a new tab)

  6. Feature design and lexicon reduction for efficient offline handwriting recognition

    … features of Arabic words such as the number of subwords and diacritics, but totally ignore the shape of the subwords. Therefore, our first goal is to perform lexicon reductionn based on subwords shape. Our approach is based on shape indexing, where the shape of a query subword is compared to a …

    ets-quebec Download PDF for Feature design and lexicon reduction for efficient offline handwriting recognition (opens in a new tab)

  7. Maximality and Applications of Subword-Closed Languages

    … its subword closure, S⊗, the set of words whose subwords of that fixed length are all in the constraint. We investigate these maximal languages and present characterizations for them. These characterizations use strongly connected components of deterministic finite automata and lead to polynomial …

    unf Repository record for Maximality and Applications of Subword-Closed Languages (opens in a new tab) Download PDF for Maximality and Applications of Subword-Closed Languages (opens in a new tab)

  8. Algorithmic Problems in Automorphic Orbits of Free Groups

    … guarantee the answer is no if they appear as subwords of the input word. Since we know some nice, isomorphism-invariant properties of one-relator groups where the relator is positive (that is, only written in positive generators), we'll also address the problem of recognizing words that can be …

    cuny-grad Repository record for Algorithmic Problems in Automorphic Orbits of Free Groups (opens in a new tab)

  9. Extensions of the Pascal Triangle to Words, and Related Counting Problems

    … coefficients of words, which count scattered subwords. The red thread of this thesis is precisely the combination of the Pascal triangle and binomial coefficients of words. The first part is dedicated to extensions of the Pascal triangle to various sets of words (languages) associated with …

    liege Repository record for Extensions of the Pascal Triangle to Words, and Related Counting Problems (opens in a new tab) Download PDF for Extensions of the Pascal Triangle to Words, and Related Counting Problems (opens in a new tab)

  10. Patterns in Words Related to DNA Rearrangements

    … a study of patterns in words that appear as subwords of words. We say that a pattern appears in a word if each pattern variable can be morphically mapped to a factor in the word. To gain insight into the complexity of, and similarities between, words, we define pattern indices and distances …

    usf Repository record for Patterns in Words Related to DNA Rearrangements (opens in a new tab)

  11. The Role of negative information when learning dense word vectors

    … for breaking up words into smaller units (subwords) – an important technique in predictive methods for representing rare words – can be successfully adapted to LexVec. We show that the explicit nature of LexVec – having access to the underlying cooccurrence matrix – allows us to selectively …

    brazil-ufrgs Repository record for The Role of negative information when learning dense word vectors (opens in a new tab)

  12. Space-aware subword tokenisation and complex word processing in language models

    … Turkish results arise because topically-relevant subwords in Turkish occur either at the start of a word or not, but very rarely in both positions. In the general domain, we also find that we can remove all word boundary information from sequences and retain equivalent performance, and that such …

    whiterose Download PDF for Space-aware subword tokenisation and complex word processing in language models (opens in a new tab)

  13. Identities of finite semigroups and related questions

    … monoid of the completion of W under taking subwords. We prove that the set of finite finitely based (infinitely based) monoids of the form S(W) is not closed either under direct products or under taking homomorphisms, or under taking subsemigroups. We also construct the first example of an …

    nebraska Repository record for Identities of finite semigroups and related questions (opens in a new tab)

  14. Novel word recognition and word spotting systems for offline Urdu handwriting

    … (or extracting) for handwritten Urdu subwords or connected components from text lines. We achieved a 92% correct segmentation rate for 372 text lines. In the word spotting algorithm, the candidate words were generated from the segmented connected components. These candidate words were …

    concordia Download PDF for Novel word recognition and word spotting systems for offline Urdu handwriting (opens in a new tab)

  15. Plurality and modification in Mandarin nominal phrases

    … that de-less modifiers are not phrasal. They are subwords in the sense of Embick and Noyer (2001). I suggest that de-less elements and nouns form morphosyntactic words, which function as subconcept terms. Hence, they are not allowed to move out of the N head and appear in the pre-D position. In …

    texas Repository record for Plurality and modification in Mandarin nominal phrases (opens in a new tab)

  16. Properties of Graphs Used to Model DNA Recombination

    … irreducible parts, i.e., its double occurrence subwords. In the final chapter we consider constructions of double occurrence words that recognize certain values for nesting index and genus range. In general, we find that for arbitrary values of nesting index [ge] 2 and genus range, there is a …

    usf Repository record for Properties of Graphs Used to Model DNA Recombination (opens in a new tab)

  17. Forbidding and Enforcing of Formal Languages, Graphs, and Partially Ordered Sets

    … of languages (fe-families) based on forbidden subwords and enforced words. This paper considers a metric on languages and proves that the metric space obtained is homeomorphic to the Cantor space. This work studies Chomsky classes of families as subspaces and shows they are neither closed nor …

    usf Repository record for Forbidding and Enforcing of Formal Languages, Graphs, and Partially Ordered Sets (opens in a new tab)

  18. Modern subword-based models for automatic speech recognition

    … extensive experiments using are done using subwords, with different types of language models including n-gram models and NNLMs. These experiments are performed on six different languages setting the new best-published result for any of these datasets. Overall, we show that subword-based …

    aalto Repository record for Modern subword-based models for automatic speech recognition (opens in a new tab)

  19. Improving very large vocabulary language modeling and decoding for speech recognition in morphologically rich languages

    … problems. Also, if all concatenations of subwords are allowed, the recognition vocabulary is unlimited. This thesis concentrates on different approaches where a limited but very large recognition vocabulary is used. This type of recognizer can, in addition to the subword-based language …

    aalto Repository record for Improving very large vocabulary language modeling and decoding for speech recognition in morphologically rich languages (opens in a new tab)

  20. Multilingual representations and models for improved low-resource language processing

    … such as words is not trivial. Although using subwords helps the models to have better coverage over unseen data and new words in the vocabulary, generalizing over low-resource languages with different alphabets and grammars is still a challenge. This thesis investigates methods to overcome …

    lmu-germany Repository record for Multilingual representations and models for improved low-resource language processing (opens in a new tab)

Page 1 of 2