TY - JOUR AU - Sárosi, Gellért AU - Tarján, Balázs AU - Fegyó, Tibor AU - Mihajlik, Péter TI - Automated transcription of conversational Call Center speech–with respect to non-verbal acoustic events JF - INTELLIGENT DECISION TECHNOLOGIES J2 - INTELL DECIS TECHNOL VL - 8 PY - 2014 IS - 4 SP - 265 EP - 275 PG - 11 SN - 1872-4981 DO - 10.3233/idt-140195 UR - https://m2.mtmt.hu/api/publication/2695587 ID - 2695587 LA - English DB - MTMT ER - TY - CHAP AU - Tarján, Balázs AU - Sárosi, Gellért AU - Fegyó, Tibor AU - Mihajlik, Péter ED - Burileanu, Corneliu ED - Teodorescu, Horia-Nicolai ED - Rusu, Corneliu TI - Improved Recognition of Hungarian Call Center Conversations T2 - 2013 7th Conference on Speech Technology and Human - Computer Dialogue (SpeD) PB - Institute of Electrical and Electronics Engineers (IEEE) CY - Piscataway (NJ) SN - 9781479910632 PY - 2013 PG - 6 DO - 10.1109/SpeD.2013.6682652 UR - https://m2.mtmt.hu/api/publication/2695417 ID - 2695417 N1 - Department of Telecommunications and Media Informatics, Budapest University of Technology and Economics, Hungary AITIA International Inc., Hungary THINKTech Research Center, Nonprofit LLC, Hungary Cited By :6 Export Date: 9 February 2023 LA - English DB - MTMT ER - TY - CHAP AU - Sárosi, Gellért AU - Tarján, Balázs AU - András, Balog AU - Tamás, Mozsolics AU - Mihajlik, Péter AU - Fegyó, Tibor TI - On Modeling Non-word Events in Large Vocabulary Continuous Speech Recognition T2 - 3rd International Conference on Cognitive Infocommunications (CogInfoCom) PB - IEEE CY - Piscataway (NJ) SN - 9781467351874 PY - 2012 SP - 649 EP - 653 PG - 5 DO - 10.1109/CogInfoCom.2012.6421932 UR - https://m2.mtmt.hu/api/publication/2689931 ID - 2689931 N1 - Budapest University of Technology and Economics, Budapest, Hungary THINKTech Research Center Nonprofit LLC, Hungary Aitia International Inc., Hungary Cited By :6 Export Date: 9 February 2023 Correspondence Address: Sárosi, G.; Budapest University of Technology and Economics, Budapest, Hungary; email: sarosi@tmit.bme.hu LA - English DB - MTMT ER - TY - CONF AU - Sárosi, Gellért AU - Fegyó, Tibor AU - Mihajlik, Péter AU - Tarján, Balázs AU - Judit, Pancza AU - Zoltán, Hans TI - LVCSR-based Speech Analytics of a Hungarian Language Call-Center T2 - Workshop on Innovation and Applications in Speech Technology PY - 2012 UR - https://m2.mtmt.hu/api/publication/2683927 ID - 2683927 LA - English DB - MTMT ER - TY - JOUR AU - Sárosi, Gellért AU - T, Mozsolics AU - Tarján, Balázs AU - A, Balog AU - Mihajlik, Péter AU - Fegyó, Tibor TI - Recognition of Multiple Language Voice Navigation Queries in Traffic Situations JF - LECTURE NOTES IN COMPUTER SCIENCE J2 - LNCS VL - 6800 PY - 2011 SP - 199 EP - 213 PG - 15 SN - 0302-9743 DO - 10.1007/978-3-642-25775-9_20 UR - https://m2.mtmt.hu/api/publication/2666039 ID - 2666039 LA - English DB - MTMT ER - TY - CHAP AU - Sárosi, Gellért AU - Mozsáry, M AU - Mihajlik, Péter AU - Fegyó, Tibor ED - Corneliu, Burileanu ED - Horia-Nicolai, Teodorescu TI - Comparison of Feature Extraction Methods for Speech Recognition in Noise-Free and in Traffic Noise Environment T2 - 2011 6th Conference on Speech Technology and Human-Computer Dialogue (SpeD) PB - IEEE CY - Piscataway (NJ) SN - 9781457704390 PY - 2011 SP - 1 EP - 8 PG - 8 DO - 10.1109/SPED.2011.5940729 UR - https://m2.mtmt.hu/api/publication/2666038 ID - 2666038 N1 - AB - A crucial part of a speech recognizer is the acoustic feature extraction, especially when the application is intended to be used in noisy environment. In this paper we investigate several novel front-end techniques and compare them to multiple baselines. Recognition tests were performed on studio quality wide band recordings on Hungarian as well as on narrow band telephone speech including real-life noises collected in six languages: English, German, French, Italian, Spanish and Hungarian. The following baseline feature types were used with several settings: Mel Frequency Cepstral Coefficients (MFCC), Perceptual Linear Prediction (PLP) features implemented in HTK, SPHINX, or by ourselves. Novel methods include Perceptual Minimum Variance Distortionless Response (PMVDR) and multiple variations of the Power-Normalized Cepstral Coefficients (PNCC). Also, adaptive techniques are applied to reduce convolutive distortions. We have experienced a significant difference between the MFCC implementations, and there were major differences in the PNCC variations useful in the different bandwidths and noise conditions. LA - English DB - MTMT ER - TY - CHAP AU - Sárosi, Gellért AU - Tobler, Z AU - Mihajlik, Péter AU - Fegyó, Tibor ED - Tanács, Attila ED - Vincze, Veronika TI - Lényegkiemelő módszerek összehasonlítása közlekedési zajban történő beszédfelismerés céljából T2 - VII. Magyar Számítógépes Nyelvészeti Konferencia : MSZNY 2010 PB - Szegedi Tudományegyetem Informatikai Tanszékcsoport CY - Szeged SN - 9789633060759 PY - 2010 SP - 384 EP - 388 PG - 5 UR - https://m2.mtmt.hu/api/publication/2666036 ID - 2666036 N1 - Besorolás: Konferenciaközlemény LA - Hungarian DB - MTMT ER - TY - JOUR AU - Sárosi, Gellért AU - Mihajlik, Péter AU - Tobler, Z AU - Fegyó, Tibor TI - Hallásmodellek a gépi beszédfelismerésben. Akusztikai lényegkiemelő módszerek összehasonlítása többnyelvű, közlekedési zajjal terhelt beszédfelismerési feladatban TS - Akusztikai lényegkiemelő módszerek összehasonlítása többnyelvű, közlekedési zajjal terhelt beszédfelismerési feladatban JF - AKUSZTIKAI SZEMLE J2 - AKUSZTIKAI SZEMLE VL - 10 PY - 2010 IS - 3-4 SP - 48 EP - 55 PG - 8 SN - 1419-6301 UR - https://m2.mtmt.hu/api/publication/2666035 ID - 2666035 LA - Hungarian DB - MTMT ER - TY - JOUR AU - Tóth, László AU - Tarján, Balázs AU - Sárosi, Gellért AU - Mihajlik, Péter TI - Speech Recognition Experiments with Audiobooks JF - ACTA CYBERNETICA J2 - ACTA CYBERN-SZEGED VL - 19 PY - 2010 IS - 4 SP - 695 EP - 713 PG - 19 SN - 0324-721X UR - https://m2.mtmt.hu/api/publication/1436367 ID - 1436367 N1 - Research Group on Artificial Intelligence, Hungarian Academy of Sciences, University of Szeged, Hungary Department of Telecommunications and Media Informatics, Budapest University of Technology and Economics, Hungary THINKTech Research Center, Nonprofit LLC, Hungary Cited By :6 Export Date: 9 February 2023 CODEN: ACCYD Correspondence Address: Tóth, L.; Research Group on Artificial Intelligence, Hungary; email: tothl@linf.u-szeged.hu LA - English DB - MTMT ER -