Language models for spontaneous speech recognition

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F15%3A00230233" target="_blank" >RIV/68407700:21230/15:00230233 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Language models for spontaneous speech recognition

Popis výsledku v původním jazyce

The paper presents the creation of n-gram Language Models (LMs) for the purposes of spontaneous speech recognition with the special focus on the recognition performed with the data from the Nijmegen Corpus of Casual Czech (NCCCz). Required LMs which cover spontaneous or casual speech re- spectively were created using the data available in the col- lected corpus NCCCz. These models were combined with a general model built on the basis Czech National Corpus. The creation of LMs, their adaptation to specific thematic do- main (casual speech) were created using SRILM toolkit. The quality of created LMs was measured on the basis Perplex- ity and Out-Of-Vocabulary words at the text level, same as they was used within the automatic speech recognition and theachieved Word Error Rates are also presented. The study shows that created LMs were suitable for given purpose and achieved results were related to other works presenting the spontaneous speech recognition.

Název v anglickém jazyce

Language models for spontaneous speech recognition

Popis výsledku anglicky

The paper presents the creation of n-gram Language Models (LMs) for the purposes of spontaneous speech recognition with the special focus on the recognition performed with the data from the Nijmegen Corpus of Casual Czech (NCCCz). Required LMs which cover spontaneous or casual speech re- spectively were created using the data available in the col- lected corpus NCCCz. These models were combined with a general model built on the basis Czech National Corpus. The creation of LMs, their adaptation to specific thematic do- main (casual speech) were created using SRILM toolkit. The quality of created LMs was measured on the basis Perplex- ity and Out-Of-Vocabulary words at the text level, same as they was used within the automatic speech recognition and theachieved Word Error Rates are also presented. The study shows that created LMs were suitable for given purpose and achieved results were related to other works presenting the spontaneous speech recognition.

Druh

D - Stať ve sborníku

CEP obor

JA - Elektronika a optoelektronika, elektrotechnika

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2015

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the 19th International Scientific Student Conferenece POSTER 2015

ISBN

978-80-01-05499-4

ISSN

—

e-ISSN

—

Počet stran výsledku

4

Strana od-do

1-4

Název nakladatele

Czech Technical University in Prague

Místo vydání

Praha

Místo konání akce

Praha

Datum konání akce

14. 5. 2015

Typ akce podle státní příslušnosti

EUR - Evropská akce

Kód UT WoS článku

—