Přejít k obsahu


First Insight into the Processing of the Language Consulting Center Data

Citace:
ZAJÍC, Z., ZAJÍCOVÁ, L., PSUTKA, J., SALAJKA, P., NOVOTNÝ, J., PRAŽÁK, , MÜLLER, L. First Insight into the Processing of the Language Consulting Center Data. In Speech and Computer 20th International Conference, SPECOM 2018 Leipzig, Germany, September 18–22, 2018, Proceedings. Cham: Springer Nature Switzerland AG, 2018. s. 778-787. ISBN: 978-3-319-99578-6 , ISSN: 0302-9743
Druh: STAŤ VE SBORNÍKU
Jazyk publikace: eng
Anglický název: First Insight into the Processing of the Language Consulting Center Data
Rok vydání: 2018
Místo konání: Cham
Název zdroje: Springer Nature Switzerland AG
Autoři: Ing. Zbyněk Zajíc Ph.D. , Ing. Lucie Zajícová Ph.D. , Ing. Mgr. Josef Psutka Ph.D. , Ing. Petr Salajka , Ing. Jaromír Novotný , Ing. Aleš Pražák Ph.D. , Doc. Ing. Luděk Müller Ph.D. ,
Abstrakt CZ: Tento článek popisuje první kroky probíhajícího projektu, jehož cílem je zpřístupnit data jazykové poradny jak pro širší veřejnost, tak i pro odborné pracovníky. Nahrané telefonní dotazy (spolu s elektronickými dopisy) je třeba poloautomaticky zpracovat a uložit v elektronické databázi pro pozdější snadné prohledávání. K tomuto účelu jsme natrénovali ASR a metody detekce témat (tzv. topic detection). V tomto článku pak prezentujeme první výsledky na reálných datech z jazykové poradny.
Abstrakt EN: In this paper, we describe the initial stages of the project “Access to a Linguistically Structured Database of Enquiries from the Language Consulting Center”. This project is attempting to provide an improved access to the large archives of mainly telephone conversations collected continuously by the Institute of the Czech Language. The main goal is to open up the unique Czech data acquired from the queries to the Language Consulting Center and to build the semi-automatic system that will facilitate searching and categorizing of these queries. For this purpose, the Automatic Speech Recognizer (ASR) and the language processing methods are being designed. The vocabulary used in such queries contains many unusual words unlike the common speech (e.g. linguistic terms). In order to train the ASR system, it is necessary to manually transcribe a large amount of speech data, identify the appropriate vocabulary, and obtain relevant text for language modeling purposes. In this paper, the proposed telephone system for recording the new data and the baseline speech recognition on these data is described. The first experiments with the topic detection on these data aimed at discovering what can be found in them and also how to preprocess them is also described.
Klíčová slova

Zpět

Patička