Rule-based Language Modeling

Notes:

Rule-based language modeling is a approach to natural language processing that involves using a set of pre-defined rules and heuristics to generate or understand human language. In contrast to statistical language modeling, which involves training a model on large amounts of data to learn the underlying patterns and structure of human language, rule-based language modeling relies on a set of pre-defined rules and heuristics that are manually defined by developers.

In the context of dialog systems, rule-based language modeling can be used to generate responses to user input. This might involve defining a set of rules that specify how the dialog system should respond to different types of user input, such as specific keywords or phrases, or particular types of questions. The system can then use these rules to generate appropriate responses to the user’s input, without needing to rely on statistical models or large amounts of training data.

One of the key advantages of rule-based language modeling is that it allows developers to have precise control over the behavior of the dialog system. By defining the rules and heuristics that the system should use, developers can ensure that the system responds in the way that they intend, and can easily modify or update the rules as needed. This can be particularly useful in applications where the dialog system needs to adhere to specific rules or constraints, or where the available data is limited or noisy.

Finite state network (FSN) is a type of mathematical model that consists of a set of states and transitions between them. Each state represents a particular situation or condition, and transitions between states are triggered by certain events or inputs. FSNs are used to represent and analyze systems with a limited number of states and transitions, and are often used in computer science and engineering to model the behavior of computer systems, communication networks, and other complex systems.
Hierarchical tree modeling approach is a method of organizing and analyzing data in the form of a tree structure, where the root node represents the overall system or concept being studied, and the child nodes represent more specific sub-concepts or components. This approach allows for a hierarchical representation of data, with higher-level concepts being broken down into more specific sub-concepts as needed.
Integrated language models are machine learning models that have been trained on a large dataset of natural language text and are able to generate human-like text by predicting the next word in a sequence based on the context of the previous words. These models are trained to capture the structure and patterns of language, and are often used for tasks such as machine translation, text generation, and language understanding.
Morpheme lexicon is a database or list of morphemes, which are the smallest units of meaning in a language. A morpheme lexicon may include information about the meaning and usage of each morpheme, as well as any rules or patterns for combining morphemes to form words. A morpheme lexicon is often used in natural language processing and linguistics research to analyze and understand the structure of language.
Phrasal patterns are sequences of words that typically occur together in a language and convey a particular meaning or function. These patterns can include collocations, idioms, and other fixed expressions that are commonly used in a language. For example, in English, the phrase “make up” can be used as a phrasal verb meaning “to invent or invent something,” as in “She made up an excuse for being late.”
Semantic constraint is a term used in linguistics and natural language processing to refer to the rules or principles that govern the meaning of words and phrases in a language. These constraints can include syntactic rules, which specify the way words can be combined to form phrases and sentences, and semantic rules, which specify the meaning of words and how they can be used in different contexts. Semantic constraints help to determine the meaning of a word or phrase based on its context and the words that surround it, and can help to disambiguate words that have multiple meanings.

Incorporating language constraints in sub-word based speech recognition H Erdo?an, O Büyük, K Oflazer – … Speech Recognition and …, 2005 – ieeexplore.ieee.org … lattices. 1. INTRODUCTION Hidden Markov Models are used in modern speech recognition systems for acoustic modelling. Statistical or rule-based language models can be used in order to represent the recognized language. If … Cited by 32 Related articles All 2 versions

A rule-based language model for reading recognition. J Cheng, B Townshend – SLaTE, 2009 – eee.bham.ac.uk Abstract Systems for assessing and tutoring reading skills place unique requirements on underlying ASR technologies. Most responses to a “read out loud” task can be handled with a low perplexity language model, but the educational setting of the task calls for diagnostic … Cited by 7 Related articles All 6 versions

A rule-based language model for speech recognition T Kaufmann – 2009 – tik.ee.ethz.ch However, n-grams fail to capture many dependencies that are present in natural language. In our experimental data, the n-gram language model prefers the incorrect transcription “einer, der sich für die Umwelt einsetzen”. This transcription is ungrammatical because the … Cited by 3 Related articles All 8 versions

Stability and accuracy in incremental speech recognition EO Selfridge, I Arizmendi, PA Heeman… – Proceedings of the …, 2011 – dl.acm.org … Two useful but different forms of language models are commonly used in spoken dialogue systems. A Rule-based Language Model (RLM) specifies a list of valid sentences which may be recognized, usually via expansion rules. … Cited by 24 Related articles All 20 versions

Estimating probability of correctness for ASR N-Best lists JD Williams, S Balakrishnan – … of the SIGDIAL 2009 Conference: The …, 2009 – dl.acm.org … acous- tic models for high-quality audio. Configuration B uses a rule-based language model consisting of all city/state pairs in the USA, and acoustic models for telephone-quality audio. Configuration A was ap- plied to corpora … Cited by 15 Related articles All 23 versions

Adaptive user preference modeling and its application to in-flight entertainment H Liu, B Salem, M Rauterberg – … of the 3rd international conference on …, 2008 – dl.acm.org … delivery systems could fall into two categories: (1) hierarchical tree modeling approach where the user preference is modeled by dimensions (eg, sports, reading), each dimension can be further refined with sub dimensions [5]; (2) rule-based language modeling approach where … Cited by 11 Related articles All 8 versions

Comparison of Grammar-Based and Statistical Language Models Trained on the Same Data BA Hockey, M Rayner – Proceedings of the AAAI Workshop on …, 2005 – researchgate.net … While these factors biased us in favor of the rule-based language modeling approach that the Clarissa system uses for speech recognition, we wanted to implement a frame- work which would allow us to compare our methods against more standard approaches based on class … Cited by 6 Related articles All 4 versions

Integrating incremental speech recognition and pomdp-based dialogue systems EO Selfridge, I Arizmendi, PA Heeman… – Proceedings of the 13th …, 2012 – dl.acm.org … We ran incremental speech recognition on each utterance using Lattice-Aware Incremental Speech Recognition (Selfridge et al., 2011) on the AT&T WATSONSM speech recognizer (Goffin et al., 2005) with the same rule-based language models used in the production system. … Cited by 15 Related articles All 26 versions

Towards accurate recognition for children’s oral reading fluency J Cheng, J Shen – Spoken Language Technology Workshop ( …, 2010 – ieeexplore.ieee.org … are discussed in detail. We show that using a set of linguis- tic rules learned from a collection of transcriptions, the pro- posed rule-based language model [1] outperformed traditional n-gram language models. Combined with a … Cited by 7 Related articles All 3 versions

Effectiveness of combined features for machine learning based question classification M Skowron, K Araki – ??????, 2005 – jlc.jst.go.jp … 65 Page 4. Journal of Natural Language Processing Vol.12 No.6 Nov.2005 3 Previous Method The approaches to question classification can be discriminated into the following three main groups:rule-based,language modeling and machine learning based1. … Cited by 11 Related articles All 5 versions

In-flight entertainment system: State of the art and research directions H Liu – Semantic Media Adaptation and Personalization, …, 2007 – ieeexplore.ieee.org … these applications could fall into two categories: (1) hierarchical tree modeling approach where the user is modeled by dimensions (eg, knowledge, or interest), each dimension can be further refined with sub dimensions [20]; (2) rule-based language modeling approach where … Cited by 10 Related articles All 8 versions

Generic and reflective graph transformations for the checking and enforcement of modeling guidelines C Amelunxen, E Legros, A Schürr – Visual Languages and …, 2008 – ieeexplore.ieee.org … The new version of SDM is the only visual rule-based language (model transformation language) we are aware of that offers both support for the visual declarative definition of generic as well as reflective transformation operations. … Cited by 8 Related articles All 6 versions

The Application of Ancient Chinese language Models in the Studies of Ancient Works [J] YHC Hua – Journal of City Polytechnic Zhuhai, 2008 – en.cnki.com.cn … language model in the studies of Chinese ancient works. ?Key Words?? statistical language models rule-based language models ancient works machine translation ?CateGory Index?? H1 ?DOI?? CNKI:SUN:ZHCZ.0.2008-04-009. … Cited by 2 Related articles

A survey on user profile modeling for personalized service delivery systems H Liu, B Salem, M Rauterberg – Proceeding of IADIS International …, 2009 – Citeseer … For the dynamic information elements such as user’s context-aware preference items which depend on the context of use, they are usually modeled with rule-based language modeling approach where the delivery of services relates to the context of use with if-then logic. … Cited by 5 Related articles All 5 versions

A Mongolian speech recognition system based on HMM G Gao, S Zhang – Computational Intelligence, 2006 – Springer … And we generate the decision tree in this way. 4 Language Model Language model consists of rule-based language model and statistical language model. By training language model, we could get the syntax and semantic parsing. … Cited by 4 Related articles All 6 versions

A Framework for Mixed-language Text-to-speech Synthesis M Malcangi, P Grew – Proceedings of the 8th WSEAS International …, 2009 – researchgate.net … 2 System Framework The system framework consists of three main components: the rule-based language model, the multi-language phonetic data, and the formant-based speech synthesizer (Fig.1). Fig. 1 – System framework for a multi-language text-to- speech synthesizer … Cited by 2 Related articles All 3 versions

Performance of automated scoring for children’s oral reading R Downey, D Rubin, J Cheng, J Bernstein – … on Innovative Use of NLP for …, 2011 – dl.acm.org … human scores. A novel aspect of the present study is that text-independent rule-based language models were employed (Cheng and Townshend, 2009) to score reading passages that the system had never seen before. Oral … Cited by 4 Related articles All 13 versions

Toward language-independent text-to-speech synthesis M Malcangi, P Grew – WSEAS Transactions on Information Science and …, 2010 – Citeseer … parentheses. 3 System Framework The system framework, as illustrated in figure 1, consists of three main components: the rule-based language model, the multi-language phonetic data, and the formant-based speech synthesizer. … Cited by 3 Related articles All 5 versions

Design of a Modified P300 Speller System Based on Prediction by Partial Matching Language Model M Wang – 2012 – rave.ohiolink.edu Page 1. Page 2. Design of a Modified P300 Speller System Based on Prediction by Partial Matching Language Model A thesis submitted to the Division of Research and Advanced Studies of the University of Cincinnati in partial fulfillment of the requirements for the degree of … Cited by 2 Related articles All 3 versions

Effectiveness of combined features for machine learning based question classification M Skowron, K Araki – Information and Media Technologies, 2006 – jlc.jst.go.jp … 464 3 Previous Method The approaches to question classification can be discriminated into the following three main groups: rule-based, language modeling and machine learning based 1 . In the rule based approach, hand-written grammar rules and a set of regular expression … Cited by 4 Related articles All 3 versions

Time scale modification and vocal tract length normalization for improving the performance of Tamil speech recognition system implemented using language … S Saraswathi, TV Geetha – International Journal of Speech Technology, 2006 – Springer Page 1. Int J Speech Technol (2006) 9: 151–163 DOI 10.1007/s10772-007-9004-y Time scale modification and vocal tract length normalization for improving the performance of Tamil speech recognition system implemented using language independent segmentation algorithm … Cited by 2 Related articles All 6 versions

Validation of automated scoring of oral reading J Balogh, J Bernstein, J Cheng… – Educational and …, 2012 – epm.sagepub.com Advertisement. … Cited by 4 Related articles All 6 versions

Survey on Swedish language resources K Elenius, E Forsbom, B Megyesi – Speech, Music and Hearing, …, 2008 – researchgate.net Page 1. Survey on Swedish Language Resources Kjell Elenius 1 Eva Forsbom 2 Beáta Megyesi 2 1 Speech, Music and Hearing School of Computer Science and Communication, KTH Sweden 2 Department of Linguistics and Philology Uppsala University Sweden 2008-03-14 … Cited by 5 Related articles All 10 versions

Voice Search–An Introduction Y Wang, D Yu, Y Ju, A Acero – IEEE Signal Processing …, 2008 – research.microsoft.com … 59.8%. Language Modeling: Early DA systems compiled directory entries into a finite-state grammar as the language model for ASR. This rule-based language model does not scale up well with directory size due to increased perplexity. … Cited by 2 Related articles All 2 versions

Rayner Manny, Ann Hockey Beth, Bouillon Pierrette, Putting Linguistics into Speech Recognition. The Regulus Grammar Compiler. With contributions by Nikos … J Bos – Natural Language Engineering, 2008 – Cambridge Univ Press … The book by Rayner, Hockey and Bouillon introduces and advocates a radically different approach: rule-based language modelling, employing a linguistically motivated grammar to produce accurate models based on context-free grammars. … Cited by 1 Related articles All 3 versions

Continuously predicting and processing barge-in during a live spoken dialogue task E Selfridge, I Arizmendi, P Heeman… – Proceedings of the …, 2013 – sigdial.org … Both systems used the AT&T WATSONSM speech recognizer (Goffin et al., 2005) with the same sub-task specific rule-based language models and standard echo cancellation techniques. The beam width was set to maximize accuracy while still running faster than real-time. … Cited by 7 Related articles All 14 versions

Syntactic language modeling with formal grammars T Kaufmann, B Pfister – Speech Communication, 2012 – Elsevier It has repeatedly been demonstrated that automatic speech recognition can benefit from syntactic information. However, virtually all syntactic language models f. Cited by 2 Related articles All 8 versions

Named entity recognition for Tibetan texts using case-auxiliary grammars H Yu, T Jiang, N Ma – Proceedings of International Multi-conference of …, 2010 – iaeng.org … a fixed list of person name. Then combining the name lexicon and the feature of the boundary words, we propose a rule-based language model for person name recognition in Tibetan texts. A. Name Lexicon There are three … Cited by 1 Related articles All 3 versions

Sub-word language modeling for Turkish speech recognition O Büyük – 2005 – busim.ee.boun.edu.tr … Hidden Markov Models are used in modern speech recognition systems for acoustic modelling. Statistical or rule-based language models can be used in order to represent the recognized language. If the speech recognition application permits the use … Cited by 1 Related articles All 3 versions

Efficient development of grammars for multilingual rule-based applications ME Santaholma – 2010 – archive-ouverte.unige.ch Page 1. Thesis Reference Efficient development of grammars for multilingual rule-based applications SANTAHOLMA, Marianne Elina Abstract Les applications de traitement automatique des langues (TALN) fondées sur des … Cited by 1 Related articles All 2 versions

Smart communication adviser for remote users M Penhaker, O Krejcar, M Cerny, M Behan… – Context-Aware Systems …, 2013 – Springer … It is based on fuzzy rule-based language model, in which the formalization of linguistic terms (linguistic values of input and output linguistic variables) using fuzzy sets to represent knowledge about how to troubleshoot the application of rules of type IF- THEN, for operation over … Cited by 2 Related articles All 4 versions

A dynamic P300-based BCI speller using a language model A Ralescu, M Wang, S Chaganti… – International Journal of …, 2014 – inderscienceonline.com … Language modelling has been used in various aspects such as automatic speech recognition, machine translation and others. Language models can be divided into two categories, rule-based language models (Brill, 1992), and statistical language models. … Related articles All 3 versions

Open Microphone Speech Understanding: Correct Discrimination of in Domain Speech J Hieronymus, G Aist, J Dowding – Acoustics, Speech and …, 2006 – ieeexplore.ieee.org … 2. RULE BASED LANGUAGE MODELS Most of the present dialogue systems use either n-gram grammars or hand built finite state grammars for speech recognition. N-gram grammars require large amounts of transcribed speech data in order to train accurate models. … Related articles All 2 versions

Executive Summary-Efficient Text Entry. K Harbusch, KJ Räihä, K Tanaka-Ishii – Efficient Text Entry, 2005 – drops.dagstuhl.de … expanded into complete sentences). At their core, these methods rely on statistical and to a lesser extent rule-based language models to predict and complete the user input and thus save keystrokes. Unlike in speech recognition … All 2 versions

Relation discovery by named entity recognition from Tibetan websites H Yu, T Jiang, B Zhang, X Chen – Web Society, 2009. SWS’09. …, 2009 – ieeexplore.ieee.org … So combining the lexicon contains 3530 [5] the most commonly used Tibetan person name and the case-auxiliary words information, we propose a rule-based language model to recognize person name from Tibetan Web pages. … Related articles All 3 versions

Effect of Speech Rate on the Speech Recognition Accuracy: A Review NU Rani, PN Girija – Artificial Intelligent Systems and Machine …, 2010 – ciitresearch.org … 89-94, 2002. Petti Vayrynen, Johannes Peltola and Tapio Seppanen, “Enhancing Phoneme Recognizer performance with a Simple Rule-based Language model”, In Proceedings of STeP- Finnish Artificial Intelligence Days, pp. 171-178, 2000. … Related articles All 2 versions

Positive Results on the LIMSI Broadcast News Data T Kaufmann – 2007 – tik.ee.ethz.ch Page 1. Positive Results on the LIMSI Broadcast News Data Interim Report for SNF Project 105211-112133: Rule-Based Language Model for Speech Recognition Tobias Kaufmann Institut für Technische Informatik und Kommunikationsnetze July 2007 Page 2. Contents … Related articles All 3 versions

Multilingual Dialogue Systems RLC Delgado, M Araki – Spoken, Multilingual and Multimodal … – Wiley Online Library … Summary; Chapter. Get PDF (3430K). Keywords: rule-based language modelling; Pronunciation Lexicon Specification (PLS); Machine Translation (MT) systems; interlingua-based architecture; semantic frame conversion architecture …

Machine Learning Based Question Classification M Skowronl, K Arakil – Journal of …, 2005 – arakilab.media.eng.hokudai.ac.jp … 2002). 65 Page 4. Journal of Natural Language Processing Vol. 12 No. 6 Nov. 2005 3 Previous Method The approaches to question classi?cation can be discriminated into the following three main groups: rule-based, language modeling and machine learning basedl. … Related articles All 2 versions

Mining the whole set of person names from the Tibetan Web T Jiang, H Yu – … Technology, 2009. ICCSIT 2009. 2nd IEEE …, 2009 – ieeexplore.ieee.org … maintain a fixed list of person name. So combining the lexicon and the case-auxiliary words information, we propose a rule-based language model for person name extraction from Tibetan Web. The algorithm is as follow: • Step … Related articles All 5 versions

Application of adaptive Chinese segmentation in chat robot Z Lv – Cyber Technology in Automation, Control, and …, 2015 – ieeexplore.ieee.org … of language model . This is a good solution to the ambiguity of the parts. Flowchart mixed Chinese rule-based language model and segmentation method is as follows: ,, Rule-based model A. Build Dictionary Since January …

Experiments on the LIMSI Broadcast News Data T Kaufmann – 2007 – tik.ee.ethz.ch Page 1. Experiments on the LIMSI Broadcast News Data Interim Report for SNF Project 105211-112133: Rule-Based Language Model for Speech Recognition Tobias Kaufmann Institut für Technische Informatik und Kommunikationsnetze February 2007 Page 2. … Related articles

Ethem Alpaydin. Introduction to Machine Learning (Adaptive Computation and Machine Learning Series). The MIT Press, 2004. ISBN: 0 262 01211 1 Price£ 32.95/ … S Khan – Natural Language Engineering, 2008 – Cambridge Univ Press … The book by Rayner, Hockey and Bouillon introduces and advocates a radically different approach: rule-based language modelling, employing a linguistically motivated grammar to produce accurate models based on context-free grammars. … Related articles All 3 versions

Reviewer S Khan, N Stannard, J Greijn, J Collins – Cambridge Univ Press … The book by Rayner, Hockey and Bouillon introduces and advocates a radically different approach: rule-based language modelling, employing a linguistically motivated grammar to produce accurate models based on context-free grammars. … All 2 versions

Different Approaches (Uszkoreit, 2001) S Müller – acat-project.eu … http://www.sx.ac.uk/linguistics/external/LFG/ www-lfg.stanford.edu/pubs/papers/kaplan/ kaplan-1989-0000.pdf , 09.04.2013. Kaufmann, Tobias. 2009. A Rule-based Language Model for Speech Recognition. Ph. D.thesis, Computer Engineering … Related articles All 2 versions

Hierarchical language modeling for one-stage stochastic interpretation of natural speech. M Thomae – 2006 – mediatum.ub.tum.de … 66 5 Hierarchical Language Models (HLM) 69 5.1 Mathematical Formulation . . . . . 71 5.2 Rule-Based Language Modeling . . . . . 74 5.3 Data-Driven Language Modeling . . . . . … Related articles All 7 versions

Estimation of Question Types for Indonesian Question Sentence A Purwarianti, T Masatoshi, S Nakagawa – must.cu-tokyo.ac.jp … 1. Introduction Question classification is an important phase in most question answering systems. There are three approaches for question classification[2]: rule based, language modeling and machine learning based. High … Related articles All 7 versions

Deep neural network acoustic models for spoken assessment applications J Cheng, X Chen, A Metallinou – Speech Communication, 2015 – Elsevier In this paper, we investigate the effectiveness of applying deep neural network hidden Markov models, or DNN-HMMs, for acoustic modeling in the context of educa.

Language Modeling for Speech Recognition Incorporating Probabilistic Topic Models MA Haidar – 2014 – espace.inrs.ca … There is a great deal of variability in natural language. Natural language grammar can be taught by two approaches: rule-based language models and statistical language models. In rule-based language models, grammar is defined as a set of rules that are accurate, but … Cited by 1 Related articles

Efficient Semantic Constraint for Speech Recognition Y Niimi, Y Kobayashiy – 130.54.20.150 … concept hierarchy was not used. 1 Introduction In this paper we present a method for describing semantic constraint in a speech under- standing system using the rule-based language model. By the rule-based language model … Related articles All 2 versions

State of the art report: in-flight entertainment system H LUI – Second International Workshop on Semantic Media …, 2007 – seat.id.tue.nl … these applications could fall into two categories: (1) hierarchical tree modeling approach where the user is modeled by dimensions (eg, knowledge, or interest), each dimension can be further refined with sub dimensions [20]; (2) rule-based language modeling approach where … Related articles

Natural language processing for the working programmer D De Kok, H Brouwer – 2011 – researchgate.net Page 1. Natural Language Processing for the Working Programmer Daniël de Kok Harm Brouwer Page 2. Natural Language Processing for the Working Programmer Daniël de Kok Harm Brouwer Copyright © 2010, 2011 Daniël de Kok, Harm Brouwer License … Cited by 8 Related articles All 4 versions

Challenges and opportunities for state tracking in statistical spoken dialog systems: Results from two public deployments JD Williams – Selected Topics in Signal Processing, IEEE …, 2012 – ieeexplore.ieee.org Page 1. Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. This article has been accepted for publication in a future issue of this journal, but has not been fully edited. … Cited by 12 Related articles All 12 versions

Glossary and Subject Index B Endres-Niggemeyer – Semantic Mashups: Intelligent Reuse of Web …, 2013 – Springer Page 348. Glossary and Subject Index Brigitte Endres-Niggemeyer Glossary and Subject Index Use The glossary lists and explains technical terms used in the book. It works as a sub- ject index, but it comes up with short descriptions … All 2 versions

AidSLT: A Spoken Language Translation Tool for Data Collection in Humanitarian Settings A Armando – 2015 – archive-ouverte.unige.ch Page 1. Master Reference AidSLT : A Spoken Language Translation Tool for Data Collection in Humanitarian Settings ARMANDO, Alejandro Abstract This master’s dissertation is concerned with the development and evaluation …

Development and evaluation of a software for second language learning: CALL-SLT C Baur – 2010 – archive-ouverte.unige.ch … 22 3.2 Speech Recognition . . . . . 25 3.2.1 Data-driven vs. rule-based language modelling . . . . . 27 3.2.1.1 Data-driven language model . . . . . 28 3.2.1.1.1 Hidden Markov Model . . . . . 28 3.2.1.2 Rule-based language model . … Related articles All 2 versions

Voice-operated Home Automation M HANSSON, E JOHANSSON, J LUNDBERG… – 2014 – publications.lib.chalmers.se … The word lattice is then processed using a grammar in order to find the final text representation. A grammar is a rule-based language model which consists of a collection of rules that state valid word sequences in the language. … Related articles

Improving speech recognition through linguistic knowledge R Beutler – 2007 – yosemite.ee.ethz.ch Page 1. Diss. ETH No. 17039 Improving Speech Recognition through Linguistic Knowledge A dissertation submitted to the SWISS FEDERAL INSTITUTE OF TECHNOLOGY ZURICH for the degree of DOCTOR OF TECHNICAL SCIENCES presented by RENÉ BEUTLER Dipl. … Cited by 5 Related articles All 7 versions

Information state based speech recognition R Jonson – 2010 – 130.241.16.4 Page 1. Information State Based Speech Recognition Page 2. Page 3. GOTHENBURG MONOGRAPHS IN LINGUISTICS 41 Information State Based Speech Recognition Rebecca Jonson Page 4. Doctoral dissertation in Linguistics, University of Gothenburg … Related articles All 6 versions

Automatic tool for screening of cognitive impairments CFL Amorim – 2014 – inesc-id.pt Page 1. Automatic tool for screening of cognitive impairments Cristiana Filipa Lopes Amorim Thesis to obtain the Master of Science Degree in Information Systems and Computer Engineering Supervisors: Prof. Isabel Maria Martins Trancoso Prof. Alberto Abad Gareta … Cited by 1 Related articles All 2 versions

Boundary infrastructures for IBIS federation: Design rationale, implementation, and evaluation J Park – 2010 – kmi.open.ac.uk Page 1. 1 Boundary Infrastructures for IBIS Federation: Design Rationale, Implementation, and Evaluation Technical Report KMI-10-01 January 2010 Jack Park Park, J. (2010). Boundary Infrastructures for IBIS Federation: Design Rationale, Implementation, and Evaluation. … Cited by 4 Related articles All 2 versions

Automatic Speech Recognition for Non-Native Speakers TT Ping – 2008 – tel.archives-ouvertes.fr Page 1. Automatic Speech Recognition for Non-Native Speakers Tan Tien Ping To cite this version: Tan Tien Ping. Automatic Speech Recognition for Non-Native Speakers. Other. Université Joseph-Fourier – Grenoble I, 2008. English. HAL Id: tel-00294973 … Related articles All 4 versions

Weighted Finite-State Methods for Spell-Checking and Correction T Pirinen – 2014 – helda.helsinki.fi Page 1. Department of Modern Languages 2014 Weighted Finite-State Methods for Spell-Checking and Correction Tommi A Pirinen Academic dissertation to be publicly discussed, by due permission of the Fac- ulty of Arts at … Cited by 1 Related articles All 4 versions