FST (Finite State Transducers)


Finite State Transducer

See also: 

Johan Schalkwyk


Finite State Transducers AP Parkes – A Concise Introduction to Languages and Machines, 2008 – Springer As the machine makes a transition between two states (according to the input symbol it is  reading, of course), it outputs one symbol. In addition to performing recognition tasks, such a  machine can also carry out computations. The purpose of this chapter is to investigate the …

[PDF] Speech recognition with weighted finite-state transducers [PDF] from ed.ac.ukM Mohri, F Pereira… – Handbook on Speech Processing …, 2008 – wiki.inf.ed.ac.uk ABSTRACT This chapter describes a general representation and algorithmic framework for  speech recognition based on weighted finite-state transducers. These transducers provide a  common and natural representation for major components of speech recognition systems, … Cited by 37 – Related articles – View as HTML – All 8 versions

[PDF] HIERARCHICAL PHRASE-BASED TRANSLATION WITH WEIGHTED FINITE STATE TRANSDUCERS [PDF] from mt-archive.infoB Byrne – 2010 – mt-archive.info R1: S?(X, X) R2: S?(SX, SX) R3: X?(s1, said) R4: X?(s1 s2, the president said) R5:  X?(s1 s2 s3, Syrian president says) R6: X?(s2, president) R7: X?(s3, the Syrian) R8:  X?(s4, yesterday) R9: X?(s5, that) R10: X?(s6, would return) R11: X?(s6, he would … View as HTML – All 3 versions

Bayesian inference for finite-state transducers [PDF] from aclweb.orgD Chiang, J Graehl, K Knight, A Pauls… – … Technologies: The 2010 …, 2010 – dl.acm.org Abstract We describe a Bayesian inference algorithm that can be used to train any cascade  of weighted finite-state transducers on end-to-end data. We also investigate the problem of  automatically selecting from among multiple training runs. Our experiments on four … Cited by 8 – Related articles – All 15 versions

Representing and combining calendar information by using finite-state transducers J Niemi… – Proceeding of the 2009 conference on Finite- …, 2009 – dl.acm.org Abstract This paper elaborates a model for representing various types of semantic calendar  expressions (SCEs), which correspond to the disambiguated intensional meanings of  natural-language calendar phrases. The model uses finite-state transducers (FSTs) to … Cited by 3 – Related articles

[PDF] Quantification and implication in semantic calendar expressions represented with finite-state transducers [PDF] from ntu.edu.twJ Niemi… – 22nd International Conference on …, 2008 – speech.ee.ntu.edu.tw Abstract This paper elaborates a model for representing semantic calendar expressions  (SCEs), which correspond to the intensional meanings of natural-language calendar  phrases. The model uses finite-state transducers (FSTs) to mark denoted periods of time … Cited by 1 – Related articles – View as HTML – All 12 versions

Dialog management using weighted finite-state transducers C Hori, K Ohtake, T Misu, H Kashioka… – … Annual Conference of …, 2008 – isca-speech.org We are aiming to construct an expandable and adaptable dialog system which handles  multiple tasks and senses users’ intention via multiple modalities. A flexible platform to  integrate different dialog strategies and modalities is indispensable for this purpose. In this … Cited by 18 – Related articles – All 5 versions

Hierarchical phrase-based translation with weighted finite-state transducers and shallow-n grammars [PDF] from cam.ac.ukA de Gispert, G Iglesias, G Blackwood… – Computational …, 2010 – dl.acm.org Abstract In this article we describe HiFST, a lattice-based decoder for hierarchical phrase- based translation and alignment. The decoder is implemented with standard Weighted Finite- State Transducer (WFST) operations as an alternative to the well-known cube pruning … Cited by 8 – Related articles – All 13 versions

Semitic morphological analysis and generation using finite state transducers with feature structures [PDF] from amharic.orgM Gasser – Proceedings of the 12th Conference of the European …, 2009 – dl.acm.org Abstract This paper presents an application of finite state transducers weighted with feature  structure descriptions, following Amtrup (2003), to the morphology of the Semitic language  Tigrinya. It is shown that feature-structure weights provide an efficient way of handling the … Cited by 12 – Related articles – All 22 versions

OpenFst: an open-source, weighted finite-state transducer library and its applications to speech and language M Riley, C Allauzen… – … : The 2009 Annual Conference of the …, 2009 – dl.acm.org Abstract Finite-state methods are well established in language and speech processing.  OpenFst (available from www. openfst. org) is a free and open-source software library for  building and using finite automata, in particular, weighted finite-state transducers (FSTs). … Cited by 3 – Related articles – All 20 versions

Language model combination and adaptation usingweighted finite state transducers [PDF] from mirlab.orgX Liu, MJF Gales, JL Hieronymus… – … Speech and Signal …, 2010 – ieeexplore.ieee.org Abstract In speech recognition systems language model (LMs) are often constructed by  training and combining multiple n-gram models. They can be either used to represent  different genres or tasks found in diverse text sources, or capture stochastic properties of … Cited by 2 – Related articles – All 7 versions

A generalized composition algorithm for weighted finite-state transducers [PDF] from google.comC Allauzen, M Riley… – Tenth Annual Conference of …, 2009 – isca-speech.org This paper describes a weighted finite-state transducer composition algorithm that  generalizes the concept of the composition filter and presents filters that remove useless  epsilon paths and push forward labels and weights along epsilon paths. This filtering … Cited by 11 – Related articles – All 10 versions

Hierarchical phrase-based translation with weighted finite state transducers [PDF] from cam.ac.ukG Iglesias, A de Gispert, ER Banga… – Proceedings of Human …, 2009 – dl.acm.org Abstract This paper describes a lattice-based decoder for hierarchical phrase-based  translation. The decoder is implemented with standard WFST operations as an alternative to  the well-known cube pruning procedure. We find that the use of WFSTs rather than k-best … Cited by 12 – Related articles – All 19 versions

Hindi Urdu machine transliteration using finite-state transducers [PDF] from aclweb.orgMG Malik, C Boitet… – Proceedings of the 22nd …, 2008 – dl.acm.org Abstract Finite-state Transducers (FST) can be very efficient to implement inter-dialectal  transliteration. We illustrate this on the Hindi and Urdu language pair. FSTs can also be  used for translation between surface-close languages. We introduce UIT (universal … Cited by 15 – Related articles – All 17 versions

Hypothesis Preservation Approach to Scene Text Recognition with Weighted Finite-State Transducer [PDF] from icdar2011.orgT Yamazoe, M Etoh, T Yoshimura… – Document Analysis and …, 2011 – ieeexplore.ieee.org Abstract This paper shows that the use of Weighted Finite-State Transducer (WFST)  significantly eliminates large-scale ambiguity in scene text recognition, especially for  Japanese Kanji characters. The proposed method consists of two WFSTs called WFST- … Related articles – All 5 versions

A Spectral Learning Algorithm for Finite State Transducers [PDF] from upc.eduB Balle, A Quattoni… – Machine Learning and Knowledge …, 2011 – Springer Finite-State Transducers (FSTs) are a popular tool for modeling paired input-output  sequences, and have numerous applications in real-world problems. Most training  algorithms for learning FSTs rely on gradient-based or EM optimizations which can be … Cited by 1 – Related articles – All 5 versions

Transformation of modular finite state transducers SE Lucco, DE Langworthy… – US Patent 7,624,075, 2009 – Google Patents (57) ABSTRACT AQ Framework, or QFX for short, is provided for performing efficient tree  transformation in a generalized manner that achieves preservation of action semantics for  FSTs that support action information in their representations across a diverse set of types … Cited by 1 – Related articles – All 3 versions

Efficient and robust music identification with weighted finite-state transducers [PDF] from cw.com.twM Mohri, PJ Moreno… – Audio, Speech, and …, 2010 – ieeexplore.ieee.org Abstract We present an approach to music identification based on weighted finite-state  transducers and Gaussian mixture models, inspired by techniques used in large-vocabulary  speech recognition. Our modeling approach is based on learning a set of elementary … Cited by 4 – Related articles – All 21 versions

3-way composition of weighted finite-state transducers [PDF] from arxiv.orgC Allauzen… – Implementation and Applications of Automata, 2008 – Springer Composition of weighted transducers is a fundamental algorithm used in many applications,  including for computing complex edit-distances between automata, or string kernels in  machine learning, or to combine different components of a speech recognition, speech … Cited by 5 – Related articles – All 18 versions

Transformation of modular finite state transducers SE Lucco, DE Langworthy… – US Patent 7,627,541, 2009 – Google Patents AQ Framework, or QFX for short, is provided for performing efficient tree transformation in a  generalized manner that achieves preservation of action semantics for FSTs that support  action information in their representations across a diverse set of types of representations … Related articles – All 3 versions

Filters for efficient composition of weighted finite-state transducers [PDF] from nyu.eduC Allauzen, M Riley… – Implementation and Application of …, 2011 – Springer This paper describes a weighted finite-state transducer composition algorithm that  generalizes the concept of the composition filter and presents various filters that process  epsilon transitions, look-ahead along paths, and push forward labels along epsilon paths. … Cited by 2 – Related articles – All 15 versions

Efficient morphological parsing with a weighted finite state transducer [PDF] from psu.eduD Cavar, IP Jazbec… – Informatica, 2009 – freepatentsonline.com This article describes a highly optimized algorithm and implementation of a deterministic  weighted finite state transducer for morphological analysis. We show how various  functionalities can be integrated into one machine, without sacrificing performance or … Cited by 1 – Related articles – Cached – All 4 versions

Ocr post-processing using weighted finite-state transducers [PDF] from icpr2010.orgR Llobet, JR Cerdan-Navarro… – … (ICPR), 2010 20th …, 2010 – ieeexplore.ieee.org Abstract A new approach for Stochastic Error-Correcting Language Modeling based on  Weighted Finite-State Transducers (WFSTs) is proposed as a method to post-process the  results of an optical character recognizer (OCR). Instead of using the recognized string as … Cited by 3 – Related articles – All 9 versions

Weighted finite state transducer based statistical dialog management C Hori, K Ohtake, T Misu… – … , 2009. ASRU 2009. …, 2009 – ieeexplore.ieee.org Abstract We proposed a dialog system using a weighted finite-state transducer (WFST) in  which user concept and system action tags are input and output of the transducer,  respectively. The WFST-based platform for dialog management enables us to combine … Cited by 5 – Related articles

Large-scale statistical machine translation with weighted finite state transducers G Blackwood, A De Gispert, J Brunning… – Proceeding of the 2009 …, 2009 – dl.acm.org Abstract The Cambridge University Engineering Department phrase-based statistical  machine translation system follows a generative model of translation and is implemented by  the composition of component models of translation and movement realised as Weighted … Cited by 5 – Related articles – All 3 versions

European language translation with weighted finite state transducers: The CUED MT system for the 2008 ACL workshop on SMT G Blackwood, A De Gispert, J Brunning… – Proceedings of the Third …, 2008 – dl.acm.org Abstract We describe the Cambridge University Engineering Department phrase-based  statistical machine translation system for Spanish-English and French-English translation in  the ACL 2008 Third Workshop on Statistical Machine Translation Shared Task. The CUED … Cited by 4 – Related articles – All 3 versions

[PDF] BASQUE-SPANISH-ENGLISH MACHINE TRANSLATION USING FINITE-STATE TRANSDUCERS [PDF] from upm.esA Pérez, MI Torres, V Guijarrubia… – 2008 – lorien.die.upm.es ABSTRACT The goals of this work are, on the one hand to develop a trilingual corpus  suitable for example-based text and speech-input machine translation between English,  Spanish and Basque; and on the other hand, to compare the same translation method … Related articles – View as HTML – All 5 versions

Disfluency detection for a speech-to-speech translation system using phrase-level machine translation with weighted finite state transducers SR Maskey, Y Gao, B Zhou – US Patent 7,860,719, 2010 – Google Patents A computer-implemented method for creating a disfluency translation lattice includes  providing a plurality of weighted finite state transducers including a translation model, a  language model, and a phrase segmentation model as input, performing a cascaded … Related articles – All 4 versions

Learning finite state transducers using bilingual phrases J González, G Sanchis… – Computational Linguistics and …, 2008 – Springer Statistical Machine Translation is receiving more and more attention every day due to the  success that the phrase-based alignment models are obtaining. However, despite their  power, state-of-the-art systems using these models present a series of disadvantages that … Cited by 7 – Related articles – BL Direct – All 6 versions

The Parametrized Probabilistic Finite-State Transducer Probe Game Player Fingerprint Model J Tsang – Computational Intelligence and AI in Games, IEEE …, 2010 – ieeexplore.ieee.org Abstract Fingerprinting operators generate functional signatures of game players and are  useful for their automated analysis independent of representation or encoding. The theory  for a fingerprinting operator which returns the length-weighted probability of a given move … Cited by 1 – Related articles – All 5 versions

Evolving finite state transducers to interpret deepwater reservoir depositional environments [PDF] from mun.caT Yu, D Wilkinson, J Clark… – … Computation, 2008. CEC …, 2008 – ieeexplore.ieee.org Abstract Predicting oil recovery efficiency of deepwater reservoirs is a challenging task. One  approach to characterize and predict the producibility of a reservoir is by analyzing its  depositional information. In a deposition-based stratigraphic interpretation framework, one … Cited by 1 – Related articles – All 3 versions

[PDF] Symbolic Finite State Transducers: Algorithms and Applications [PDF] from microsoft.comM Veanes, P Hooimeijer, B Livshits… – Proceedings of the …, 2012 – research.microsoft.com Abstract Finite automata and finite transducers are used in a wide range of applications in  software engineering, from regular expressions to specification languages. We extend these  classic objects with symbolic alphabets represented as parametric theories. Admitting … Cited by 1 – View as HTML

[CITATION] Adrià de Gispert, Eduardo R. Banga, William Byrne, Hierarchical phrase-based translation with weighted finite state transducers G Iglesias – Proceedings of Human Language Technologies: The …, 2009 Cited by 7 – Related articles

[PDF] Linguistic Categorisation in Machine Translation using Stochastic Finite State Transducers [PDF] from mt-archive.infoJ González… – Mixing Approaches to Machine …, 2008 – mt-archive.info Abstract In the last years, statistical machine translation has already demonstrated its  usefulness within a wide variety of translation applications. In particular, finite state models  are always an interesting framework because there are well-known efficient algorithms for … Cited by 1 – Related articles – View as HTML

Exploiting Morphology in Speech Translation with Phrase-Based Finite-State Transducers [PDF] from psu.eduA Pérez, MI Torres… – … Languages Technologies for …, 2010 – isca-speech.org This work implements a novel formulation for phrase-based translation models making use  of morpheme-based translation units under a stochastic finite-state framework. This  approach has an additional interest for speech translation tasks since it leads to the … Related articles – All 5 versions

A Stochastic Finite-State Transducer Approach to Spoken Dialog Management LF Hurtado, J Planells, E Segarra… – … Annual Conference of …, 2010 – isca-speech.org In this paper, we present an approach to spoken dialog management based on the use of a  Stochastic Finite-State Transducer estimated from a dialog corpus. The states of the  Stochastic Finite-State Transducer represent the dialog states, the input alphabet includes … Cited by 3 – Related articles

[PDF] Symbolic finite state transducers: Algorithms and applications [PDF] from microsoft.comN Bjorner, P Hooimeijer, B Livshits, D Molnar… – month, 2011 – research.microsoft.com Finite automata and finite transducers are used in a wide range of applications in software  engineering, from regular expressions to specification languages. We extend these classic  objects with symbolic alphabets represented as parametric theories. Admitting potentially … Cited by 1 – Related articles – View as HTML – All 2 versions

[PDF] Hidden semi-Markov Model based earthquake classification system using Weighted Finite-State Transducers [PDF] from nonlin-processes-geophys.netM Beyreuther, J Wassermann – Nonlin. Processes …, 2011 – nonlin-processes-geophys.net Abstract. Automatic earthquake detection and classification is required for efficient analysis  of large seismic datasets. Such techniques are particularly important now because access to  measures of ground motion is nearly unlimited and the target waveforms (earthquakes) … Cited by 1 – Related articles – View as HTML – All 3 versions

[PDF] Decision procedures for composition and equivalence of symbolic finite state transducers [PDF] from microsoft.comM Veanes, B Livshits… – month, 2011 – research.microsoft.com ABSTRACT Finite automata model a wide array of applications in software engineering,  from regular expressions to specification languages. Finite transducers are an extension of  finite automata to model functions on lists of elements, which in turn have uses in fields as … Cited by 2 – Related articles – View as HTML – All 2 versions

Guessers for Finite-State Transducer Lexicons [PDF] from helsinki.fiK Lindén – Computational Linguistics and Intelligent Text …, 2009 – Springer Language software applications encounter new words, eg, acronyms, technical terminology,  names or compounds of such words. In order to add new words to a lexicon, we need to  indicate their inflectional paradigm. We present a new generally applicable method for … Cited by 5 – Related articles – All 5 versions

A Weighted Finite-State Transducer (WFST)-based Language Model for Online Indic Script Handwriting Recognition [PDF] from icdar2011.orgS Chowdhury, U Garain… – Document Analysis and …, 2011 – ieeexplore.ieee.org Abstract Though designing of classifies for Indic script handwriting recognition has been  researched with enough attention, use of language model has so far received little  exposure. This paper attempts to develop a weighted finite-state transducer (WFST) based … Related articles – All 3 versions

[CITATION] General Topics in Speech Recognition-Learning a Discriminative Weighted Finite-State Transducer for Speech Recognition M Lehr… – IEEE Transactions on Audio Speech and …, 2011

[CITATION] A new Epsilon Filter for Efficient Composition of Weighted Finite-State Transducers F Duckhorn, M Wolff… – 2011 Related articles

Silence models in weighted finite-state transducers [PDF] from 192.33.221.237PN Garner – Ninth Annual Conference of the International Speech …, 2008 – isca-speech.org We investigate the effects of different silence modelling strategies in Weighted Finite-State  Transducers for Automatic Speech Recognition. We show that the choice of silence models,  and the way they are included in the transducer, can have a significant effect on the size of … Cited by 5 – Related articles – All 13 versions

A discriminative model for continuous speech recognition based on weighted finite state transducers [PDF] from mirlab.orgS Watanabe, T Hori, E McDermott… – Acoustics Speech and …, 2010 – ieeexplore.ieee.org Abstract This paper proposes a discriminative model for speech recognition that directly  optimizes the parameters of a speech model represented in the form of a decoding graph. In  the process of recognition, a decoder, given an input speech signal, searches for an … Cited by 1 – Related articles – All 4 versions

Automatic Sanskrit segmentizer using finite state transducers [PDF] from aclweb.orgV Mittal – Proceedings of the ACL 2010 Student Research …, 2010 – dl.acm.org Abstract In this paper, we propose a novel method for automatic segmentation of a Sanskrit  string into different words. The input for our segmentizer is a Sanskrit string either encoded  as a Unicode string or as a Roman transliterated string and the output is a set of possible … Cited by 4 – Related articles – All 13 versions

A multiplatform speech recognition decoder based on weighted finite-state transducers [PDF] from kit.eduE Stoimenov… – Automatic Speech Recognition & …, 2009 – ieeexplore.ieee.org Abstract Speech recognition decoders based on static graphs have recently proven to  significantly outperform the traditional approach of prefix tree expansion in terms of  decoding speed. The reduced search effort makes static graph decoders an attractive … Related articles – All 5 versions

Finite state transducers with intuition R Agadzanyan… – Unconventional Computation, 2010 – Springer Finite automata that take advice have been studied from the point of view of what is the  amount of advice needed to recognize nonregular languages. It turns out that there can be  at least two different types of advice. In this paper we concentrate on cases when the … Cited by 1 – Related articles – All 3 versions

TRANSFORMATION OF MODULAR FINITE STATE TRANSDUCERS SE Lucco, DE Langworthy… – EP Patent …, 2009 – freepatentsonline.com … TRANSFORMATION OF MODULAR FINITE STATE TRANSDUCERS. European Patent Application EP2062132 Kind Code: A1. Inventors: Lucco, Steven E. (One Microsoft Way, Redmond, Washington 98052-6399, US) Langworthy, David E. (One Microsoft Way, Redmond …  Cached

TRANSFORMATION OF MODULAR FINITE STATE TRANSDUCERS SE Lucco, DE Langworthy… – EP Patent …, 2009 – freepatentsonline.com … TRANSFORMATION OF MODULAR FINITE STATE TRANSDUCERS. European Patent Application EP2062133 Kind Code: A2. Inventors: Lucco, Steven E. (One Microsoft Way, Redmond, Washington 98052-6399, US) Langworthy, David E. (One Microsoft Way, Redmond …  Cached

[CITATION] Adrià de Gispert, Jamie Brunning, William Byrne, European language translation with weighted finite state transducers: the CUED MT system for the 2008 … G Blackwood – Proceedings of the Third Workshop on Statistical …, 2008 Cited by 2 – Related articles

Application of Weighted Finite-State Transducers to Improve Recognition Accuracy for Dysarthric Speech O Caballero Morales… – 2008 – ueaeprints.uea.ac.uk Abstract Standard speaker adaptation algorithms perform poorly on dysarthric speech  because of the limited phonemic repertoire of dysarthric speakers. In a previous paper, we  proposed the use of” metamodels” to correct dysarthric speech. Here, we report on an … Cached

A DNA-based Finite State Transducer [PDF] from cnrs.frN Aubert – 2010 – dumas.ccsd.cnrs.fr Deoxyribonucleic acid (DNA) computing is an emerging eld wherein the biological  properties of DNA are used to carry out computation. As DNA can also be used as a building  material with movable parts, the combination of those two aspects allows scientists to … Related articles – All 6 versions

European language translation with weighted finite state transducers: the CUED MT system GW Blackwood, A de Gispert… – … of the ACL 2008 …, 2008 – publications.eng.cam.ac.uk Blackwood, GW and de Gispert, A and Brunning, JJJ and Byrne, WJ (2008) European language  translation with weighted finite state transducers: the CUED MT system. In: ACL 2008 Third Workshop  on Statistical Machine Translation, 19-6-2008 to –, Columbus, Ohio pp. 131-135..  Cached

[HTML] Syntactic error detection and correction in date expressions using finite-state transducers [HTML] from google.comADD Ilarraza, K Gojenola, M Oronoz… – Finite-state methods …, 2008 – books.google.com Abstract. This paper presents a system for the detection and correction of syntactic errors. It  combines a robust morphosyntactic analyser and two groups of ?nite-state transducers  speci?ed using the Xerox Finite State Tool (xfst). One of the groups is used for the … Related articles – All 5 versions

GREAT: a finite-state machine translation toolkit implementing a Grammatical Inference Approach for Transducer Inference (GIATI) [PDF] from uni-potsdam.deJ González… – Proceedings of the EACL 2009 Workshop …, 2009 – dl.acm.org … re- ported. In (González and Casacuberta, 2007), the so called phrase-based finite-state transducers were concluded to be a better modelling option for this task than the ones that derive from a word- based approach. That is …  Cited by 4 – Related articles – All 17 versions

Systems and methods for determining the determinizability of finite-state automata and transducers C Allauzen… – US Patent 7,783,485, 2010 – Google Patents … cited by examiner Primary Examiner-Matthew J Sked (57) ABSTRACT Finite-state transducers and weighted finite-state automata may not be determinizable. … For a weighted or unweighted finite-state transducer, that device is checked to determine if it is functional. …  Related articles – All 5 versions

Flexible and Expandable Speech Recognition Hardware with Weighted Finite State Transducers K You, J Choi… – Journal of Signal Processing Systems, 2011 – Springer Abstract Hardware implementation of speech recognition can not only accelerate decoding  speed for real-time processing but also reduce the power consumption. Recently the  weighted finite state transducer (WFST) has emerged as a major recognition network … Related articles

An FPGA implementation of speech recognition with weighted finite state transducers [PDF] from mirlab.orgJ Choi, K You… – Acoustics Speech and Signal …, 2010 – ieeexplore.ieee.org Abstract In this paper we present a hardware architecture for large vocabulary continuous  speech recognition that conducts a search over a weighted finite state transducer (WFST)  network. A pipelined architecture is proposed to fully utilize the memory bandwidth. A … Cited by 1 – Related articles – All 4 versions

Polynomial time identification of strict prefix deterministic finite state transducers [PDF] from upv.esM Wakatsuki… – Grammatical Inference: Theoretical Results and …, 2010 – Springer This paper is concerned with a subclass of finite state transducers, called strict prefix  deterministic finite state transducers (SPDFST’s for short), and studies a problem of  identifying the subclass in the limit from positive data. After providing some properties of … Related articles – All 4 versions

A Weighted Finite-State Transducer Implementation of Phoneme Rewrite Rules for English to Korean Pronunciation Conversion H Koo – Procedia-Social and Behavioral Sciences, 2011 – Elsevier Words change their phonetic as well as orthographic form when they are borrowed and  used by speakers of another language. A formal model that properly captures this change  has theoretical implications in phonology and practical applications in speech processing …

Designing antimicrobial peptides with weighted finite-state transducers [PDF] from sonmezsysbio.orgC Whelan, B Roark… – Engineering in Medicine …, 2010 – ieeexplore.ieee.org Abstract The design of novel antimicrobial peptides (AMPs) is an important problem given  the rise of drug-resistant bacteria. However, the large size of the sequence search space,  combined with the time required to experimentally test or simulate AMPs at the molecular … Related articles – All 3 versions

Part-of-speech tagging using parallel weighted finite-state transducers [PDF] from helsinki.fiM Silfverberg… – Advances in Natural Language Processing, 2010 – Springer We use parallel weighted finite-state transducers to implement a part-of-speech tagger,  which obtains state-of-the-art accuracy when used to tag the Europarl corpora for Finnish,  Swedish and English. Our system consists of a weighted lexicon and a guesser combined … Cited by 1 – Related articles – All 5 versions

[PDF] An Open-Source Finite State Morphological Transducer for Modern Standard Arabic [PDF] from rug.nlM Attia, P Pecina, A Toral, L Tounsi… – … Workshop Finite State …, 2011 – acl.eldoc.ub.rug.nl … the singular in- definite form. Other inflected forms are generated from the lemma through alteration rules. In our implementation of the Arabic finite state transducer, we use the lemma as the base form. We believe that a lemma …  View as HTML

[PDF] A Rule based Kannada Morphological Analyzer and Generator using Finite State Transducer [PDF] from ijcaonline.orgV Ramasamy, PJ Antony, S Saravanan… – International Journal of …, 2011 – ijcaonline.org Abstract Morphology plays an essential role in machine translation and many other natural  language processing applications. Developing a well fledged morphological analyzer and  generator (MAG) tools for highly agglutinative language like Kannada is a challenging … Related articles – View as HTML – All 2 versions

Application of Weighted Finite-State Transducers to Improve Recognition Accuracy for Dysarthric Speech [PDF] from uea.ac.ukOC Morales… – Ninth Annual Conference of the International …, 2008 – isca-speech.org Standard speaker adaptation algorithms perform poorly on dysarthric speech because of the  limited phonemic repertoire of dysarthric speakers. In a previous paper, we proposed the  use of” metamodels” to correct dysarthric speech. Here, we report on an improved … Related articles – All 3 versions

On the statistical estimation of stochastic finite-state transducers in machine translation J Andrés-Ferrer, F Casacuberta-Nolla… – Applied Artificial …, 2008 – Taylor & Francis The inference of finite-state transducers from bilingual training data plays an important role  in many natural-language tasks and mainly in machine translation. However, there are only  a few techniques to infer such models. One of these techniques is the grammatical … Cited by 2 – Related articles – BL Direct – All 3 versions

[PDF] Standardization of Terms Applying Finite-State Transducers (FST) [PDF] from ugr.esC Galvez – Handbook of research on digital libraries: design, …, 2009 – ugr.es ABSTRACT This chapter presents the different standardization methods of terms at the two  basic approaches of nonlinguistic and linguistic techniques, and sets out to justify the  application of processes based on finitestate transducers (FST). Standardization of terms … Related articles – View as HTML – All 12 versions

Finite-state transducers and speech recognition in Slovak language M Lojka… – Signal Processing Algorithms, Architectures, …, 2009 – ieeexplore.ieee.org Abstract In speech recognition systems based on statistical methods is possible to transform  the speech recognition problem into the finding the most probable path through a speech  recognition network (SRN). The path through the SRN identifies sequence of recognized … Cited by 1 – Related articles

[PDF] Designing a Non-Finite-State Weighted Transducer Toolkit [PDF] from columbia.eduSF Chen – 2009 – ee.columbia.edu … 7 Related Work Obviously, the largest influence of this work is the AT&T finite-state transducer library (Mohri et al., 1998), which has had an immense impact on the field of language processing and others. … 2002. Parameter estimation for probabilistic finite-state transducers. …  Cited by 1 – Related articles – All 4 versions

[PDF] General Algorithms for Testing the Ambiguity of Finite Automata and the Double-Tape Ambiguity of Finite-State Transducers [PDF] from nyu.eduC ALLAUZEN, M MOHRI… – International Journal of …, 2011 – cs.nyu.edu A finite automaton is ambiguous if it admits distinct accepting paths with the same label. The  question of the ambiguity of finite automata arises in a variety of contexts. In some cases, the  application of an algorithm requires an input automaton to be finitely ambiguous, in others, … Cited by 1 – Related articles – View as HTML – All 5 versions

A Compiler for Morphological Analyzer Based on Finite-State Transducers BC Melinamath, AG Math… – Information Systems for Indian …, 2011 – Springer Morphological analyzers are an essential parts of many natural language processing (NLP)  systems such as machine translation systems. They may be efficiently implemented as finite  state transducers. This paper describes a morphological system that can be used as … Related articles – All 2 versions

Information extraction from semi-structured resources: a two-phase finite state transducers approach V Pajic, G Lažetic… – Implementation and Application of Automata, 2011 – Springer The paper presents a new method for extracting information from semi-structured resources,  based on finite state transducers. The method has two clearly distinguished phases. The first  phase-pre-processing phase-strongly relies upon the analysis of the document structure … Related articles – All 3 versions

Low space-complexity composition algorithm for weighted finite-state transducers W LI, J WU… – Application Research of Computers, 2011 – en.cnki.com.cn The WFST-related composition algorithm could be used to integrate recognition models  together to facilitate the utilization of knowledge during speech recognition and to improve  the recognition system’s performance. The general composition algorithm stores lots of … Cached

Arabic language modeling with finite state transducers [PDF] from aclweb.orgI Heintz – Proceedings of the 46th Annual Meeting of the …, 2008 – dl.acm.org Abstract In morphologically rich languages such as Arabic, the abundance of word forms  resulting from increased morpheme combinations is significantly greater than for languages  with fewer inflected forms (Kirchhoff et al., 2006). This exacerbates the out-of-vocabulary ( … Cited by 4 – Related articles – All 14 versions

Chinese-English transliteration using weighted finite-state transducers P Wei… – Audio, Language and Image Processing, 2008. …, 2008 – ieeexplore.ieee.org Abstract This paper proposes a novel method for Chinese-English transliteration based on  multiple models by using weighted finite-state transducers (WFST). WFST provide a unified  framework for integrating the various components of a speech-to-speech translation … Related articles

[PDF] Putting Encyclopaedia Knowledge into Structural Form: Finite State Transducers Approach [PDF] from imbio.deV Pajic – Journal of Integrative Bioinformatics, 2011 – journal.imbio.de Summary In biology and functional genomics in particular, understanding the dependence  and interplay between different genome and ecological characteristics of organisms is a  very challenging problem. There are some public databases which combine this kind of … Related articles – View as HTML – All 3 versions

[HTML] SynCoP: combining syntactic tagging with chunking using weighted finite state transducers [HTML] from google.comJ Didakowski – Finite-state methods and natural language …, 2008 – books.google.com Abstract. This paper describes the key aspects of the system SynCoP (Syntactic Constraint  Parser) developed at the Berlin-Brandenburgische Akademie der Wissenschaften. The  parser allows to combine syntactic tagging and chunking by means of constraint grammar … Cited by 3 – Related articles – All 6 versions

Proper Noun Recognition and Classification Using Weighted Finite State Transducers J Didakowski… – Proceeding of the 2009 conference on …, 2009 – dl.acm.org Abstract This paper presents a new approach to proper noun recognition and classification  in which the knowledge of ambiguities within morphological analyses is used exhaustively  in the analysis. Here a proper noun recognizer/classifier is defined by proper noun context … Related articles

Part-of-speech tagging and partial parsing for Irish using finite-state transducers and constraint grammar [PDF] from dcu.ieE Uí Dhonnchadha – 2009 – doras.dcu.ie Abstract In this thesis, we present the development and evaluation of a suite of annotation  tools for unrestricted Irish text, which go from tokenization, morphological analysis, part-of- speech tagging, right through to partial parsing. In order to develop such tools, a large … Cited by 1 – Related articles – All 2 versions

[CITATION] Finite State Transducers in Web Monitoring V Pajic – Master Thesys, Faculty of Mathematics, University of …, 2010 Cited by 2 – Related articles

Joining composition and trimming of finite-state transducers J Bubenzer… – Finite-State Methods and Natural Language …, 2010 – Springer The composition of two (weighted) finite-state transducers is usually carried out in two steps:  In the first step, all accessible states of the result are constructed regardless of their co- accessibility. Non-co-accessible states are removed afterwards in the second step. This … Related articles – All 3 versions

[PDF] Finite State Transducer Modification by Examples [PDF] from jku.atG Guta – 2010 – risc.jku.at Abstract Model Transformation By Example (MTBE) is a new branch of model driven  software development. Transducers (automata with output) can be used to abstract model  transformation. This form of abstraction makes possible to consider applications of … Cited by 1 – Related articles – View as HTML – All 4 versions

CroMo-Croatian Morphology and Finite State Transducer: Morphological analysis of Croatian (and other languages) D Cavar – 2008 – bib.irb.hr … Other. Authors: Cavar, Damir. Title: CroMo – Croatian Morphology and Finite State Transducer: Morphological analysis of Croatian (and other languages). Source: na mrežnim stranicama, kod i dokumentacija. … Year: 2008. Keywords: Finite State Transducer, morphology, parser. …  Cached – All 3 versions

Realisierung von Mischverteilungsdichten durch gewichtete Automaten (Finite-State Transducer) F Duckhorn, M Wolff… – ITG-Fachbericht- …, 2010 – vde-verlag.de Abstract: Ziel dieser Arbeit ist es, Möglichkeiten zur Verwendung von gewichteten,  endlichen, sequenziellen Automaten (weighted Finite-State Transducer) zur Realisierung  von Mischverteilungsdichten in Hidden-Markov-Modellen zu untersuchen. Solche … Related articles – All 2 versions

A morphological generator for Kannada based on finite state transducers BC Melinamath… – … Technology (ICECT), 2011 …, 2011 – ieeexplore.ieee.org Abstract We present morphological generator for Kannada Language. Our work is novel in  that it explicitly addresses the need for processing the morphology of the language. This  system performs generation of inflected forms given root+ pattern+ features representation … Related articles

Towards the Development of an Automatic Diacritizer for the Persian Orthography based on the Xerox Finite State Transducer [PDF] from uottawa.caP Nojoumian – 2011 – ruor.uottawa.ca Due to the lack of short vowels or diacritics in Persian orthography, many Natural Language  Processing applications for this language, including information retrieval, machine  translation, text-to-speech, and automatic speech recognition systems need to … All 2 versions

FINITE STATE TRANSDUCER DETERMINATION DEVICE, AND FINITE STATE TRANSDUCER DETERMINATION METHOD M NAGAO – WO Patent WO/2010/116,554, 2010 – wipo.int Abstract: A non-single-valued NFST can be determined without pretreatment. Provided are a  state merging unit which extracts, from among states of the finite state transducer, a plurality  of states which transit in response to a same input signal, and merges the states; and a … Cached – All 2 versions

Speech and text translation involving basque language: an application of stochastic finite-state transducers A Perez Ramirez – 2010 – dialnet.unirioja.es Resumen: El objeto de este trabajo es la traducción automática de texto y de voz tomando  el euskera bien como lengua origen, bien como lengua destino, No nos restringimos a la  traducción de texto, sino que consideramos también la voz. Cached

Design of a finite-state transducer for parts of speech tagging of Turkish Ü Mersinli – 2010 – mitosweb.com Mitos Üyeleri. Ankara Üniversitesi Açik Erisim Sistemi; Atatürk Üniversitesi Açik Erisim Sistemi; Atilim Üniversitesi Açik Erisim Sistemi; Bahçesehir Üniversitesi Açik Erisim Sistemi; Çukurova Üniversitesi Açik Erisim Sistemi; Dokuz …  Cached

TRANSFORMATION OF MODULAR FINITE STATE TRANSDUCERS SE LUCCO, DE LANGWORTHY… – WO Patent WO/2008/ …, 2008 – wipo.int Abstract: AQ Framework, or QFX for short, is provided for performing efficient tree  transformation in a generalized manner that achieves preservation of action semantics for  FSTs that support action information in their representations across a diverse set of types … Cached – All 2 versions

Memory subsystem design for hardware speech recognition with weighted finite state transducers ???, ??? – 2010 – s-space.snu.ac.kr ??? ?????; Large Vocabulary Continuous Speech Recognition (LVCSR); ?? ??  ?? ???; Weighted Finite States Transducer (WFST); ???? ??? ??; Hardware  Viterbi search; ????? FPGA ??; FPGA implementation of speech recognition Cached