Informacja

Drogi użytkowniku, aplikacja do prawidłowego działania wymaga obsługi JavaScript. Proszę włącz obsługę JavaScript w Twojej przeglądarce.

English (EN)·Polski (PL)·Yкраїнський (UK)
Przejdź do strony domowej biblioteki Centralna Biblioteka Wojskowa Link otwiera się w nowym oknie Logo biblioteki Prolib Integro - strona głównaCentralna Biblioteka Wojskowa

Wyszukujesz frazę "language models" wg kryterium: Temat

  Lista filtrów
Wyrównaj
Wyświetlanie 1-15 z 23
Tytuł:
The Metaphor of Metalanguage from the Perspective of Contemporary Linguistics: Insights into the Relationship between Conceptualization and the Theoretical Modelling of Language, Communication and Translation
Autorzy:
Bąk, Paweł
Powiązania:
https://bibliotekanauki.pl/articles/2013184.pdf
Data publikacji:
2021
Wydawca:
Komisja Nauk Filologicznych Polskiej Akademii Nauk, Oddział we Wrocławiu
Tematy:
metalinguistic metaphorics
conduit metaphor
language models of communication and translation
Anthropological Linguistics
Cognitive Linguistics
Opis:
The paper aims at discussing the problems of conceptual metalinguistic metaphorics, its influence on the way theoretical considerations is exercised in contemporary linguistic discourse, which results in the creation of specific models of language, communication and translation. The problem of metaphorical profiling discussed in the following considerations is approached from the perspective of contemporary linguistics, which makes it possible to see a peculiar confusion of “reality” and “cognitive” aspects in metalinguistic reflection. Undoubtedly, it is impossible to think and discuss the most abstract aspects of language outside the metaphorical framework. But should all metaphorical imagery, including the most “common sense” constructs, justify the theoretical modelling of language, communication or translation and set the course for a scientific cognitive approach to these categories? The metaphorical language of the researcher should be an element consistent with the theoretical assumptions that the researcher represents. This is because a metaphor is a reflection of a person’s way of thinking. A lack of congruence between expressions and the description of the declared views may result in a lack of credibility of the language theory. The conclusions presented in this paper will be subjected to further investigations.
Źródło:
Academic Journal of Modern Philology; 2021, 14; 55-67
2299-7164
2353-3218
Pojawia się w:
Academic Journal of Modern Philology
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Homo ethnicus v zerkale jazyka: k metodike opisanija
HOMO ETHNICUS IN THE MIRROR OF LANGUAGE: TOWARDS A METHODOLOGY OF DESCRIPTION
Autorzy:
Berezovič, Elena L.
Gulik, Dimitrij P.
Powiązania:
https://bibliotekanauki.pl/articles/611436.pdf
Data publikacji:
2002
Wydawca:
Uniwersytet Marii Curie-Skłodowskiej. Wydawnictwo Uniwersytetu Marii Curie-Skłodowskiej
Tematy:
wyobrażenia o obcych
konotacja etnonimów
„portret onomazjologiczny”
wewnętrzna forma etnonimów
derywaty semantyczne
onomastyka
modele nazewnicze
język rosyjski
język angielski
images of alien ethnic groups
connotations of ethnonyms
“onomasiological portrait”
the internal structure of ethnonyms
semantic derivatives
onomasiology
naming models
the Russian language
the English language
Opis:
The article is an attempt to reconstruct images of alien ethnic groups on the basis of cultural connotations of ethnonyms. The analysis leads to a reconstruction of an „onomasiological portrait”. It consists of naming models, in which the linguistic knowledge about a given real-world object is reflected in the „internal form” of lexical items (especially in the internal form of nickname-ethnonyms and their semantic derivatives). In order lo reconstruct a relatively complete picture of a given phenomenon, it is important to take into account the lexis of different language varieties: dialects, jargons, the literary standard, as well as onomastic data. It is also important to consider the contrastive aspect of the phenomenon, to compare the portraits of the same ethnic group in different languages, which is why the present. work offers an analysis of Russian and English data. The types of information identifiable through an inquiry into naming models are characterized. Special attention is paid to the factor of linguistic technique, which may significantly alter the content of the object being reconstructed.
utorzy rozpatrują możliwości rekonstrukcji wyobrażeń o obcych etnosach na podstawie analizy kulturowej konotacji etnonimów. Analiza taka prowadzi do rekonstrukcji „portretu onomazjologicznego”. Składa się on z modeli nazewniczych, w których „wiedza językowa” o obiekcie rzeczywistości odbija się w „wewnętrznej formie” jednostek leksykalnych (mowa głównie o wewnętrznej formie etnonimów-przezwisk i tworzonych od nich derywatów semantycznych). Sprawą zasadniczej wagi jest uwzględnienie leksyki z różnych odmian języka: gwar, żargonów, języka literackiego, a także faktów z dziedziny onomastyki. Wszystko to pozwala na uzyskanie stosunkowo pełnego obrazu badanego zjawiska. Ważny jest również aspekt kontrastywny, zakładający m.in. porównanie portretów tego samego etnosu w różnych językach; w związku z tym autorzy dokonują porównawczej analizy dwóch języków: rosyjskiego i angielskiego. W artykule scharakteryzowano typy informacji, jakie można wyodrębnić w wyniku badań modeli nazewniczych. Szczególną uwagę zwracają autorzy na analizę czynnika techniki językowej, który może wnieść istotne korekty treściowe do rekonstruowanego obiektu.
Źródło:
Etnolingwistyka. Problemy Języka i Kultury; 2002, 14; 47-67
0860-8032
Pojawia się w:
Etnolingwistyka. Problemy Języka i Kultury
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
ChatGPT: Cross cultural tourism research imperative
Autorzy:
Chen, Sirong
Zhang, Ke
Li, Xiaonan
Ye, Huiyue
Lin, Katsy Jiaxin
Law, Rob
Powiązania:
https://bibliotekanauki.pl/articles/2199826.pdf
Data publikacji:
2023
Wydawca:
Uniwersytet Ekonomiczny w Katowicach
Tematy:
ChatGPT
large language models
tourism
cross-cultural analysis
Opis:
Aim/purpose – Cultural differences play an influential role in driving tourism development and varying outcomes of technology acceptance. The impact of ChatGPT on tourism is yet to be explored. Therefore, this study discussed how ChatGPT may influence tourism development and management in different cultural contexts. Design/methodology/approach – Drawing specifically on individualism and collectivism from Hofstede’s cultural dimensions theory, this study compared the differences in the perception of ChatGPT between Oriental and Western cultures. Findings – Individualism and collectivism explain the diversity of perception on ChatGPT between Oriental and Western cultures. Research implications/limitations – Implications and suggestions are provided to academia and industry accordingly. Particularly, the difference between technological concepts and tools is highlighted to avoid having excessive expectations toward tools. Originality/value/contribution – This study extends the research on tourism and Artificial Intelligence and advocates additional attention to the imperative of cross-cultural differences regarding ChatGPT and tourism.
Źródło:
Journal of Economics and Management; 2023, 45; 137-146
1732-1948
Pojawia się w:
Journal of Economics and Management
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Measuring Goal-Oriented Requirements Language Actor Stability
Autorzy:
Hassine, Jameleddine
Alshayeb, Mohammad
Powiązania:
https://bibliotekanauki.pl/articles/384139.pdf
Data publikacji:
2019
Wydawca:
Politechnika Wrocławska. Oficyna Wydawnicza Politechniki Wrocławskiej
Tematy:
goal models
Goal-oriented Requirements Language
GRL
stability
metrics
maintenance
Opis:
Background: Goal models describe interests, preferences, intentions, desired goals and strategies of intervening stakeholders during the early requirements engineering stage. When capturing the requirements of real-world systems such as socio-technical systems, the produced goal models evolve quickly to become large and complex. Hence, gaining a sufficient level of understanding of such goal models, to perform maintenance tasks, becomes more challenging. Metric-based approaches have shown good potential in improving software designs and making them more understandable and easier to maintain. Aim: In this paper, we propose a novel metric to measure GRL (Goal-oriented Requirements Language) “actor stability” that provides a quantitative indicator of the actor maintainability. Method: We first, validate the proposed metric theoretically then empirically using a case study of a GRL model describing the fostering of the relationship between the university and its alumni. Results: The proposed actor stability metric is found to have significant negative correlation with the maintenance effort of GRL models. Conclusions: Our results show that the proposed metric is a good indicator of GRL actors’ stability.
Źródło:
e-Informatica Software Engineering Journal; 2019, 13, 1; 203-226
1897-7979
Pojawia się w:
e-Informatica Software Engineering Journal
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Application of automatic speech recognition to medical reports spoken in Polish
Autorzy:
Hnatkowska, B.
Sas, J.
Powiązania:
https://bibliotekanauki.pl/articles/333379.pdf
Data publikacji:
2008
Wydawca:
Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
Tematy:
systemy informacji medycznej
modele językowe
automatic speech recognition
hospital information systems
language models
Opis:
The paper presents an attempt to automatic speech recognition of Polish spoken medical texts. The attempt resulted in experimental system that can be used as a tool for practical applications. The system uses a typical recognition method based on Hidden Markov Model and domain-specific language model. Implemented software made it possible to conduct many experiments aimed on evaluation of the assumed approach usefulness. Obtained experiment results are presented and analyzed. The system architecture and the way in which it can be integrated with hospital information systems is also exposed.
Źródło:
Journal of Medical Informatics & Technologies; 2008, 12; 223-229
1642-6037
Pojawia się w:
Journal of Medical Informatics & Technologies
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Основные модели метафорической номинации в русском и болгарском языках (на материале семантических инновантов – названий лиц)
Models of Metaphorical Nomination of Nouns Denoting Persons in the Russian and Bulgarian Languages
Autorzy:
Ivanova, Vanya
Powiązania:
https://bibliotekanauki.pl/articles/968266.pdf
Data publikacji:
2015
Wydawca:
Uniwersytet Łódzki. Wydawnictwo Uniwersytetu Łódzkiego
Tematy:
metaphorical nomination
metaphorization models
nouns denoting persons
russian and bulgarian language
метафорическая номинация
модели метафоризации
названия лиц
русский язык и болгарский язык
Opis:
The article describes the main models of metaphorical nomination of nouns denoting persons in the Russian and Bulgarian languages in the latest period. Sets out most active models of metaphorization in both languages, identifies similarities and differences in the appearance of secondary meanings based on metaphorization.
В статье рассматриваются основные модели метафорической номинации при развитии вторичных значений у существительных имен – названий лиц в русском и болгарском языках в новейший период. Установляются самые активные модели метафоризации в обоих языках, выявляются сходства и различия при возникновении новых значений на основе метафоризации.
Źródło:
Acta Universitatis Lodziensis. Folia Linguistica Rossica; 2015, 11; 37-47
1731-8025
2353-9623
Pojawia się w:
Acta Universitatis Lodziensis. Folia Linguistica Rossica
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Four b-learning models in childrens speech-language therapy
Autorzy:
Jatkowska, Joanna
Powiązania:
https://bibliotekanauki.pl/articles/426146.pdf
Data publikacji:
2019
Wydawca:
Szkoła Główna Handlowa w Warszawie
Tematy:
speech-language therapy
speech disorders
information and communication technologies
e-learning platform
b-learning
b-learning models
Opis:
B-learning is a new tool in children’s speech-language pathology, one which has not previously been tested empirically by Polish scholars. Polish speech-language pathology has used information and communication technologies for diagnostic and therapeutic purposes for more than 30 years. Originally, speech-language pathologists in Poland used computer software translated from other languages, but in the 1990s, they began to develop their own programs to accommodate the specificity of the Polish language. The advent of the Internet brought with it speech-language therapy e-learning platforms. The author of this paper created a remote speech-language therapy course concept in 2009, which was used to create the first Polish e-learning platform for children with speech impediments, available at www.elogo.edu.pl. Developed in 2010, it was part of the EU project: “Stworzenie internetowego serwisu E-Logo-Edukacja, opartego na innowacyjnych e-usługach: e-logopedia i e-statystyka” (“Development of the E-Logo-Edukacja online service based on innovative e-services: e-speech therapy and e-statistics”). The platform has been used in the speech-language therapy of children with speech impediments in order to verify if new technologies can help improve articulation. The creator of the concept named it b-learning, and it incorporates online exercises into traditional speech-language therapy. A study conducted between 2011 and 2014 on two 35-person groups (an experimental group and a control group) proved that b-learning can be effective when applied in children’s speech-language therapy. This paper demonstrates the effectiveness of four b-learning models. The results obtained proved that the most effective b-learning model involved the concurrent use of traditional speech-language therapy and the e-learning platform.
Źródło:
e-mentor. Czasopismo naukowe Szkoły Głównej Handlowej w Warszawie; 2019, 5 (82); 18-26
1731-6758
1731-7428
Pojawia się w:
e-mentor. Czasopismo naukowe Szkoły Głównej Handlowej w Warszawie
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Relacje między modelami w soteriologii. Glossa do dyskusji o komplementarności
Autorzy:
Kempa, Jacek
Powiązania:
https://bibliotekanauki.pl/articles/1368775.pdf
Data publikacji:
2020-12-10
Wydawca:
Akademia Katolicka w Warszawie
Tematy:
methodology of theology
language of theology
theological models
complementarity of the theological models
Opis:
 W artykule została zaproponowana korekta w ujmowaniu relacji między modelami soteriologicznymi. Zwykle określa się ją mianem komplementarnej. Po skrótowym przedstawieniu stanu dyskusji nad ideą idei modeli w teologii i zasadą komplementarności zwrócono uwagę na specyfikę modeli soteriologicznych: ich wielość, która wymyka się konstruowaniu dychotomicznych układów (typowych dla wzorcowej idei komplementarności i stosowalnej np. w trynitologii), a także ich niejednorodność wynikającą z niejednakowego powiązania ze źródłami wiary. Relacje między modelami w soteriologii są na tyle skomplikowane, że nie zawsze pomaga w ich wyjaśnieniu „standardowa” zasada komplementarności, choć zawsze pozostaje w mocy reguła niewyrażalności tajemnicy zbawienia przy pomocy jednego tylko wyjaśnienia, a zatem także reguła uzupełniania się różnych (poprawnych) ujęć. Pojawiające się w teologii próby syntetycznego ujęcia soteriologii nie mogą się tym regułom sprzeciwiać, ale też ich podejmowanie świadczy o tym, że można odstawiać na bok tezę o pozostawaniu poszczególnych modeli we wzajemnym, nieusuwalnym napięciu.
The article proposes a correction in the recognition of the relationships between soteriological models. Usually, the relation is referred to as complementary. After briefly presenting the state of discussion on the idea of models in theology and the principle of complementarity, attention is drawn to the specificity of soteriological models: their multiplicity, which eludes the construction of dichotomous systems (typical for the model idea of complementarity and applicable e.g. in trinitology), as well as their heterogeneity resulting from connections with sources of faith. The relationships between the models in soteriology are so complicated that the „standard” principle of complementarity does not always help in their explanation, although the rule of the inexpressibility of the mystery of salvation using only one explanation and therefore the rule of complementing different (correct) approaches always remains in force. The attempts of a synthetic approach to soteriology appearing in theology cannot oppose these rules, but the sole undertaking proves that one can put aside the thesis that individual models remain in mutual, irremovable tension.
Źródło:
Studia Bobolanum; 2020, 31, 2; 53-68
1642-5650
2720-1686
Pojawia się w:
Studia Bobolanum
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
A probabilistic model of Ancient Egyptian writing
Autorzy:
Nederhof, M.-J.
Rahman, F.
Powiązania:
https://bibliotekanauki.pl/articles/103801.pdf
Data publikacji:
2017
Wydawca:
Polska Akademia Nauk. Instytut Podstaw Informatyki PAN
Tematy:
Ancient Egyptian
writing systems
language models
Opis:
This article offers a formalization of how signs form words in Ancient Egyptian writing, for either hieroglyphic or hieratic texts. The formalization is in terms of a sequence of sign functions, which concurrently produce a sequence of signs and a sequence of phonemes. By involving a class of probabilistic automata, we can define the most likely sequence of sign functions that relates a given sequence of signs to a given sequence of phonemes. Experiments with two texts are discussed.
Źródło:
Journal of Language Modelling; 2017, 5, 1; 131-163
2299-856X
2299-8470
Pojawia się w:
Journal of Language Modelling
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
When the print meets the screen: Towards a model of L1 and L2 online reading comprehension
Autorzy:
Piasecka, Liliana Katarzyna
Powiązania:
https://bibliotekanauki.pl/articles/606014.pdf
Data publikacji:
2019
Wydawca:
Uniwersytet Marii Curie-Skłodowskiej. Wydawnictwo Uniwersytetu Marii Curie-Skłodowskiej
Tematy:
offline and online reading
models of reading
foreign/ second language reading comprehension
Opis:
Der Band enthält die Abstracts ausschließlich in englischer Sprache.
The spread of information and communication technologies (ICTs) has been changing literacy practices and activities. Consequently, the traditional view of literacy as the ability to read and write needs to be revised to encompass new forms of literacy called e-literacy, or “digital/silicon/electronic literacies” (Murray & McPherson, 2006, p. 132) as well as “hyperreading” (Usó-Juan & Ruiz-Madrid, 2009, p. 59). The members of the 21st century “global”, “fluid” and “networked” (Jewitt, 2008) societies engage in activities that the access to the WWW makes possible. The aim of the paper is to discuss print-based models of reading, identify similarities and differences between online and offline text comprehension, also with respect to the foreign/second language (L2) reading. Online texts entail the necessity to use different sets of skills and strategies which have to be incorporated into a model of electronic text comprehension. Suggestions as to what such a model might include, based on theoretical underpinnings and empirical findings, are presented.
L'article contient uniquement les résumés en anglais.
Źródło:
Lublin Studies in Modern Languages and Literature; 2019, 43, 3
0137-4699
Pojawia się w:
Lublin Studies in Modern Languages and Literature
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Three Translators in Search of an Author: Linguistic Strategies and Language Models in the (Re)translation of Shakespeare’s Plays into Catalan
Autorzy:
Pujol, Dídac
Powiązania:
https://bibliotekanauki.pl/articles/648038.pdf
Data publikacji:
2017
Wydawca:
Uniwersytet Łódzki. Wydawnictwo Uniwersytetu Łódzkiego
Tematy:
Shakespeare
translation
retranslation
adaptation
oral discourse
TV3’s oral standard
Catalan-Spanish code-switching and diglossia
Catalan literary language
linguistic strategies
language models
Gaietà Soler
Anfòs Par
Salvador Oliva
Hamlet
Kin
Opis:
This article shows how the language of Shakespeare’s plays has been rendered into Catalan in three especially significant periods: the late 19th century, the early 20th century, and the late 20th and early 21st centuries. The first section centres on the contrast between natural and unnatural language in Hamlet, and considers how this differentiation is carried out (by linguistic techniques that differ substantially from Shakespeare’s) in a late 19th-century Catalan adaptation by Gaietà Soler. The second part of the article investigates the reasons why in an early 20th-century translation of King Lear the translator, Anfòs Par, resorts to medieval instead of present-time language. The last section of the article illustrates how and explores the motivations why Salvador Oliva’s first (1985) version of The Tempest is retranslated in 2006 using a different language model. The ultimate aim of the paper is to put forward the hypothesis that, in the case of Catalan, Shakespearean translations are both a reflection of the current state of the language and a major linguistic experimentation that shapes and creates (sometimes through a via negativa) the Catalan literary language.
Źródło:
Multicultural Shakespeare: Translation, Appropriation and Performance; 2017, 16, 31; 41-59
2083-8530
2300-7605
Pojawia się w:
Multicultural Shakespeare: Translation, Appropriation and Performance
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Między uniwersytetem a szkołą. Jak przygotować studentów specjalności nauczycielskiej do współczesnego kształcenia językowego
Between school and university. How to prepare teaching specialization students for modern language education
Autorzy:
Rypel, Agnieszka
Powiązania:
https://bibliotekanauki.pl/articles/1044448.pdf
Data publikacji:
2019-12-15
Wydawca:
Uniwersytet im. Adama Mickiewicza w Poznaniu
Tematy:
knowledge transmission
contemporary models of language
education
university tasks
Opis:
The purpose of this article is to show the role of the university in the transfer of knowledge and to prepare future teachers to transfer this knowledge at school. This is an extremely important problem due to contemporary cultural and social changes causing very dynamic language changes. The methods used to describe the language are changing as well. The language education model is already anachronistic in Polish schools because it doesn’t respond to this evolutionm which is why universities must prepare future teachers to react to changes on their own.
Źródło:
Poznańskie Studia Polonistyczne. Seria Językoznawcza; 2019, 26, 2; 231-251
1233-8672
2450-4939
Pojawia się w:
Poznańskie Studia Polonistyczne. Seria Językoznawcza
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Die Modularität als Schlüssel zum Verständnis des Sprachverarbeitungssystems – kognitive Modelle und ihre neuronale Basis
Autorzy:
Sadownik, Barbara
Powiązania:
https://bibliotekanauki.pl/articles/915292.pdf
Data publikacji:
2014-01-01
Wydawca:
Uniwersytet im. Adama Mickiewicza w Poznaniu
Tematy:
modularity of mind
modular approaches
modular versus connectionist models of language processing
language perception
language production
syntactic processes in speech production
generative grammar
neuroimaging
Opis:
The scope of this paper is the concept of modularity of language, which is a key to understanding the complexity of human cognitive system. The author of the paper presents various arguments in favor of the thesis that both moderate and extreme modularity-related theories allow, too much greater extent than non-modularity theories, for reconstruction and description of complex structures of the language perception and production.
Źródło:
Glottodidactica. An International Journal of Applied Linguistics; 2014, 41, 1; 51-73
0072-4769
Pojawia się w:
Glottodidactica. An International Journal of Applied Linguistics
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Application of local bidirectional language model to error correction in polish medical speech recognition
Autorzy:
Sas, J.
Powiązania:
https://bibliotekanauki.pl/articles/333597.pdf
Data publikacji:
2010
Wydawca:
Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
Tematy:
rozpoznawanie mowy
modele języka
medyczne systemy informacji
speech recognition
language models
medical information systems
Opis:
In the paper, the method of short word deletion errors correction in automatic speech recognition is described. Short word deletion errors appear to be a frequent error type in Polish speech recognition. The proposed speech recognition process consists of two stages. At the first stage the utterance is recognized by a typical speech recognizer based on forward bigram language model. At the second stage the word sequence recognized by the first stage recognizer is analyzed and such pairs of adjacent words in the recognized sequence are localized, which are likely to be separated by a short word like conjunction or preposition. The probability of short word appearance in context of found words is evaluated using centered trigrams and backward bigram language model for short words prone to deletion. The set of probabilistic language properties used to correct deletions is called here Local Bidirectional Language Model (in contrast to purely forward or backward model used typically in speech recognition). The decision of short word insertion is based on comparison of deletion error probability of the first stage recognizer and the error probability of the decision based only on centered trigrams and backward model. Despite its simplicity, the method proved to be effective in correcting deletion errors of most frequently appearing Polish prepositions. The method was tested in application to medical spoken reports recognition, where the overall short word deletion error rate was reduced by almost 45%.
Źródło:
Journal of Medical Informatics & Technologies; 2010, 15; 127-134
1642-6037
Pojawia się w:
Journal of Medical Informatics & Technologies
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Optimal acoustic model complexity selection in polish medical speech recognition
Autorzy:
Sas, J.
Poreba, T.
Powiązania:
https://bibliotekanauki.pl/articles/333361.pdf
Data publikacji:
2011
Wydawca:
Uniwersytet Śląski. Wydział Informatyki i Nauki o Materiałach. Instytut Informatyki. Zakład Systemów Komputerowych
Tematy:
rozpoznawanie mowy
modele języka
medyczne systemy informacji
speech recognition
language models
medical information systems
Opis:
In the paper, the method of acoustic model complexity level selection for automatic speech recognition is proposed. Selection of the appropriate model complexity affects significantly the accuracy of speech recognition. For this reason the selection of the appropriate complexity level is crucial for practical speech recognition applications, where end user effort related to the implementation of speech recognition system is important. We investigated the correlation between speech recognition accuracy and two popular information criteria used in statistical model evaluation: Bayesian Information Criterion and Akaike Information Criterion computed for applied acoustic models. Experiments carried out for language models related to general medicine texts and radiology diagnostic reporting in CT and MR showed strong correlation of speech recognition accuracy and BIC criterion. Using this dependency, the procedure of Gaussian mixture count selection for acoustic model was proposed. Application of this procedure makes it possible to create the acoustic model maximizing the speech recognition accuracy without additional computational costs related to alternative cross-validation approach and without reduction of training set size, which is unavoidable in the case of cross-validation approach.
Źródło:
Journal of Medical Informatics & Technologies; 2011, 17; 115-122
1642-6037
Pojawia się w:
Journal of Medical Informatics & Technologies
Dostawca treści:
Biblioteka Nauki
Artykuł
Wyświetlanie 1-15 z 23

Ta witryna wykorzystuje pliki cookies do przechowywania informacji na Twoim komputerze. Pliki cookies stosujemy w celu świadczenia usług na najwyższym poziomie, w tym w sposób dostosowany do indywidualnych potrzeb. Korzystanie z witryny bez zmiany ustawień dotyczących cookies oznacza, że będą one zamieszczane w Twoim komputerze. W każdym momencie możesz dokonać zmiany ustawień dotyczących cookies