Dr. Tamás Gábor CSAPÓ

Address:
Department of Telecommunications and Media Informatics
Budapest University of Technology and Economics
Budapest, Hungary

Office:
H-1117 Budapest, Magyar tudósok körútja 2.
Building I, Room B.152.

Status:
Researcher

Phone:
+36-1-463-3512

Fax:
+36-1-463-3107

E-mail:
csapotATtmit.bme.hu

CV:
in English

Social networks:
ResearchGate, LinkedIn, Facebook, Google Scholar, Mendeley

Research area:
speech synthesis

Language literacy:
English - fluent
German - fluent
Hungarian - native

International activity:
2014 January - July - Indiana University, Fulbright visiting student researcher, Bloomington, IN, USA
2009 August - University of Joensuu, ECSE Summer School, Finland
2009 March - Telecom ParisTech, Athens programme, France

Lecturing:
Human-Computer Interaction
Infocommunication

Membership:
Scientific Association for Infocommunications Hungary
International Speech Communication Association
The Institute of Electrical and Electronics Engineers (IEEE) - Signal Processing Society (SPS)

PhD dissertation
Increasing the naturalness of synthesized speech in hidden Markov-model based text-to-speech synthesis
Dissertation (in Hungarian)
Summary (in Hungarian)
Summary (in English)
Public defense

Publications

Publications in the MTMT2 database


Downloading any one of these documents indicates that you agree to abide by a copyright notice.

In English


2019

·   Tamás Gábor Csapó, Mohammed Salah Al-Radhi, Géza Németh, Gábor Gosztolya, Tamás Grósz, László Tóth, Alexandra Markó, ,,Ultrasound-based Silent Speech Interface Built on a Continuous Vocoder'', Interspeech 2019, accepted. arXiv:1906.09885

·   Alexandra Markó, Márton Bartók, Tamás Gábor Csapó, Tekla Etelka Graczi, Andrea Deme, ,,Articulatory analysis of transparent vowel /i?/ in harmonic and antiharmonic Hungarian stems: Is there a difference?'', Interspeech 2019, accepted.

·   Andrea Deme, Márton Bartók, Tekla Etelka Graczi, Tamás Gábor Csapó, Alexandra Markó, ,,V-to-V coarticulation induced acoustic and articulatory variability of vowels: The effect of pitch-accent'', Interspeech 2019, accepted.

·   Andrea Deme, Márton Bartók, Tekla Etelka Gráczi, Tamás Gábor Csapó, Alexandra Markó, ,,Articulatory organization of geminates in Hungarian'', ICPhS 2019, accepted.

·   Alexandra Markó, Márton Bartók, Tamás Gábor Csapó, Andrea Deme, Tekla Etelka Gráczi, ,,The effect of focal accent on vowels in Hungarian: Articulatory and acoustic data'', ICPhS 2019, accepted.

·   Eloi Moliner Juanpere, Tamás Gábor Csapó, ,,Ultrasound-Based Silent Speech Interface Using Convolutional and Recurrent Neural Networks'', Acta Acustica united with Acustica - Fast Track, Vol. 105, pp. 587-590, 2019. paper (Open Access)

·   Mohammed Salah Al-Radhi, Tamás Gábor Csapó, Géza Németh, ,,Adaptive Refinements of Pitch Tracking and HNR Estimation within a Vocoder for Statistical Parametric Speech Synthesis'', Applied Sciences, 9 (12), 2460, 2019. paper (Open Access)

·   Dagoberto Porras, Alexander Sepúlveda-Sepúlveda, Tamás Gábor Csapó, ,,DNN-based Acoustic-to-Articulatory Inversion using Ultrasound Tongue Imaging'', IJCNN 2019, (International Joint Conference on Neural Networks), Budapest, Hungary, 2019, accepted. arXiv:1904.06083

·   Gábor Gosztolya, Ádám Pintér, László Tóth, Tamás Grósz, Alexandra Markó, Tamás Gábor Csapó, ,,Autoencoder-Based Articulatory-to-Acoustic Mapping for Ultrasound Silent Speech Interfaces'', IJCNN 2019, (International Joint Conference on Neural Networks), Budapest, Hungary, 2019, accepted. arXiv:1904.05259

·   Mohammed Salah Al-Radhi, Tamás Gábor Csapó, Géza Németh, ,,RNN-based speech synthesis using a continuous sinusoidal model'', IJCNN 2019, (International Joint Conference on Neural Networks), Budapest, Hungary, 2019, accepted. arXiv:1904.06075


2018

·   Alexandra Markó, Andrea Deme, Márton Bartók, Tekla Etelka Gráczi, Tamás Gábor Csapó, ,,Word-Initial Irregular Phonation as a Function of Speech Rate and Vowel Quality in Hungarian'', International Seminar on Speech Production, ISSP 2017: Studies on Speech Production, Tianjin, China, 2018, pp 134-145. paper

·   Mohammed Salah Al-Radhi, Tamás Gábor Csapó, Géza Németh ,,A Continuous Vocoder using Sinusoidal Model for Statistical Parametric Speech Synthesis'', SPECOM 2018, Leipzig, Germany, 2018, pp 11-20. paper

·   László Tóth, Gábor Gosztolya, Tamás Grósz, Alexandra Markó, Tamás Gábor Csapó, ,,Multi-Task Learning of Speech Recognition and Speech Synthesis Parameters for Ultrasound-based Silent Speech Interfaces'', Interspeech 2018, Hyderabad, India, 2018, pp. 3172-3176, paper,

·   Alexandra Markó, Andrea Deme, Márton Bartók, Tekla Etelka Gráczi, Tamás Gábor Csapó, ,,Speech Rate and Vowel Quality Effects on Vowel-related Word-initial Irregular Phonation in Hungarian'', CHALLENGES IN ANALYSIS AND PROCESSING OF SPONTANEOUS SPEECH, MTA Nyelvtudományi Intézet, Budapest, Hungary 2018, pp. 49-74. paper

·   Alexandra Markó, Márton Bartók, Tekla Etelka Gráczi, Andrea Deme, Tamás Gábor Csapó, ,,Prominence Effects on Hungarian Vowels: A Pilot Study'', Speech Prosody 2018, Poznan, Poland, 2018, pp. 868-872. paper

·   Tamás Grósz, Gábor Gosztolya, László Tóth, Tamás Gábor Csapó, Alexandra Markó, ,,F0 Estimation for DNN-Based Ultrasound Silent Speech Interfaces'', ICASSP 2018, Calgary, Alberta, Canada, pp. 291-295, 2018, paper, poster

·   Julia Eichholz, Michelle Meier, Reinhold Greisbach, Helma Pasch, Germain Landi, Tamás Gábor Csapó, Alexandra Markó, Andrea Deme, ,,Vocalic tongue shape contours in Zande'', Proceedings of the Conference on Phonetics & Phonology in German-speaking countries, Berlin, Germany, 2018, pp. 49-52. paper


2017

·   Mohammed Salah Al-Radhi, Tamás Gábor Csapó, Géza Németh ,,Continuous vocoder in feed-forward deep neural network based speech synthesis'', DOGS 2017, Novi Sad, Serbia, pp. 1-4, 2017. paper

·   Tamás Gábor Csapó, Andrea Deme, Tekla Etelka Gráczi, Alexandra Markó, ,,Comparison of distance measures in tongue contour traces of ultrasound images'', Ultrafest8, Potsdam, Germany, 2017. abstract poster pic

·   Tamás Gábor Csapó, Tamás Grósz, Gábor Gosztolya, László Tóth, Alexandra Markó, ,,DNN-based Ultrasound-to-Speech Conversion for a Silent Speech Interface'', Interspeech 2017, Stockholm, Sweden, pp. 3672-3676, 2017. paper presentation

·   Mohammed Salah Al-Radhi, Tamás Gábor Csapó, Géza Németh ,,Time-domain envelope modulating the noise component of excitation in a continuous residual-based vocoder for statistical parametric speech synthesis'', Interspeech 2017, Stockholm, Sweden, pp. 434-438, 2017. paper

·   Mohammed Salah Al-Radhi, Tamás Gábor Csapó, Géza Németh ,,Deep Recurrent Neural Networks in Speech Synthesis Using a Continuous Vocoder'', SPECOM 2017, Hatfield, Hertfordshire, UK, pp. 282-291, 2017. paper

·   Tamás Gábor Csapó, Andrea Deme, Tekla Etelka Gráczi, Alexandra Markó, Gergely Varjasi, ,,Synchronized speech, tongue ultrasound and lip movement video recordings with the "Micro" system'', Workshop on Challenges in Analysis and Processing of Spontaneous Speech (CAPSS2017), Budapest, Hungary, pp. 48-49, 2017. abstract

·   Kele Xu, Pierre Roussel, Tamás Gábor Csapó, Bruce Denby, ,,Convolutional neural network-based automatic classification of midsagittal tongue gestural targets using B-mode ultrasound images'', The Journal of the Acoustical Society of America - Express Letters 141 (6), EL531-EL537, 2017. link paper

·   Alexandra Markó, Tamás Gábor Csapó, Karolina Takács, ,,Listeners’ evaluation of voice quality in Hungarian speakers'', Beszédkutatás 25, pp. 55-66, 2017.


2016

·   Tamás Gábor Csapó, Géza Németh, Milos Cernak, Philip N. Garner, ,,Modeling Unvoiced Sounds In Statistical Parametric Speech Synthesis with a Continuous Vocoder'', EUSIPCO 2016 (24th European Signal Processing Conference), Budapest, Hungary, pp. 1338-1342, 2016. paper presentation pic

·   Bálint Pál Tóth, Tamás Gábor Csapó, ,,Continuous Fundamental Frequency Prediction with Deep Neural Networks'', EUSIPCO 2016 (24th European Signal Processing Conference), Budapest, Hungary, pp. 1348-1352, 2016. paper presentation

·   Kele Xu, Tamás Gábor Csapó, Pierre Roussel, Bruce Denby, ,,A comparative study on the contour tracking algorithms in ultrasound tongue images with automatic re-initialization'', The Journal of the Acoustical Society of America - Express Letters 139 (5), EL154-EL160, 2016. link

·   Milan Sečujski, Branislav Gerazov, Tamás Gábor Csapó, Vlado Delić, Philip N. Garner, Aleksandar Gjoreski, David Guennec, Zoran Ivanovski, Aleksandar Melov, Géza Németh, Ana Stojković, György Szaszák, ,,Design of a Speech Corpus for Research on Cross-Lingual Prosody Transfer'', SPECOM 2016 (18th International Conference on Speech and Computer), Budapest, Hungary, pp. 199-206, 2016. paper


2015

·   Tamás Gábor Csapó, Géza Németh, Milos Cernak, ,,Residual-based excitation with continuous F0 modeling in HMM-based speech synthesis'', SLSP 2015 (3rd International Conference on Statistical Language and Speech Processing), Budapest, Hungary, Lecture Notes in Artificial Intelligence 9449, pp. 27-38, 2015. paper presentation samples

·   Tamás Gábor Csapó, Steven M. Lulich, ,,Error analysis of extracted tongue contours from 2D ultrasound images'', Interspeech 2015, Dresden, Germany, pp. 2157-2161, 2015. paper, poster, videos, pic

·   Tamás Gábor Csapó, Géza Németh, ,,Automatic transformation of irregular to regular voice by residual analysis and synthesis'', Interspeech 2015, Dresden, Germany, pp. 613-617, 2015. paper, poster, pic

·   Kálmán Abari, Tamás Gábor Csapó, Bálint Pál Tóth, Gábor Olaszy, ,,From text to formants - indirect model for trajectory prediction based on a multi-speaker parallel speech database'', Interspeech 2015, Dresden, Germany, pp. 623-627, 2015. paper, poster, demo


2014

·   Rebecca Pedro, Elizabeth Mazzocco, Tamás G. Csapó, Steven M. Lulich, ,,Investigation of a tongue-internal coordinate system for two-dimensional ultrasound'', The Journal of the Acoustical Society of America, Vol. 136, No. 4, p. 2128. (168th Meeting of ASA, Indianapolis, IN) abstract

·   Tamás Gábor Csapó, Steven M. Lulich, ,,Comparison of tongue contour extraction methods from ultrasound images for use in text-to-speech synthesis'', Inaugural Conference of the Hungarian Cultural Association, Bloomington, IN, USA, April 6, 2014. abstract presentation pic

·   Tamás Gábor Csapó, Géza Németh, ,,Modeling irregular voice in statistical parametric speech synthesis with residual codebook based excitation'', IEEE Journal on Selected Topics in Signal Processing, Vol. 8., No. 2., pp. 209-220, 2014. link paper

·   Tamás Gábor Csapó, Géza Németh, ,,Statistical parametric speech synthesis with a novel codebook-based excitation model'', Intelligent Decision Technologies, Vol. 8., No. 4., pp. 289-299, 2014. link paper

·   György Szaszák, Tamás Gábor Csapó, Philip N. Garner, Branislav Gerazov, Zoran Ivanovski, Géza Németh, Bálint Tóth, Milan Secujski, Vlado Delic, ,,The SP2 SCOPES Project on Speech Prosody'', Digital speech and image processing (DOGS 2014), Novi Sad, Serbia, 2014. paper

·   António Teixeira, Annika Hämäläinen, Jairo Avelar, Nuno Almeida, Géza Németh, Tibor Fegyó, Csaba Zainkó, Tamás Csapó, Bálint Tóth, André Oliveira, Miguel Sales Dias, ,,Speech-centric Multimodal Interaction for Easy-to-access Online Services – A Personal Life Assistant for the Elderly'', Procedia Computer Science (5th International Conference on Software Development and Technologies for Enhancing Accessibility and Fighting Info-exclusion, DSAI 2013), Vol. 27., pp. 289-397, 2014. link paper


2013

·   Tamás Gábor Csapó, ,,Increasing the naturalness of synthesized speech in hidden Markov-model based text-to-speech synthesis'', PhD thesis, BME TMIT, 2013. dissertation (in Hungarian), summary (in Hungarian), summary (in English)

·   Tamás Gábor Csapó, Géza Németh, ,,A novel irregular voice model for HMM-based speech synthesis'', ISCA 8th Speech Synthesis Workshop (SSW8), Barcelona, Spain, pp. 229-234. link paper presentation samples pic


2012

·   Tamás Gábor Csapó, ,,Increasing the naturalness of synthesized speech (PhD summary)'', The Phonetician, No. 105-106, 2012/I-II, pp. 88-97 paper website

·   Tamás Gábor Csapó, Géza Németh, ,,A novel codebook-based excitation model for use in speech synthesis'', CogInfoCom 2012, Kosice, Slovakia, pp. 661-665. paper presentation pic video

·   Éva Székely, Tamás Gábor Csapó, Bálint Tóth, Péter Mihajlik, Julie Carson-Berndsen ,,Synthesizing Expressive Speech from Amateur Audiobook Recordings'', SLT 2012, Miami, Florida, USA, pp. 297-302. paper


2011

·   Tekla Etelka Gráczi, Steven M. Lulich, Tamás Gábor Csapó, András Beke, ,,Context and speaker dependency in the relation of vowel formants and subglottal resonances - Evidence from Hungarian'', Interspeech 2011, Florence, Italy, pp. 1901-1904. paper poster pic

·   Géza Németh, Gábor Olaszy, Tamás Gábor Csapó: ,,Spemoticons: Text-To-Speech based emotional auditory cues'', ICAD 2011, Budapest, Hungary. paper

·   Tamás Gábor Csapó, Tekla Etelka Gráczi, Zsuzsanna Bárkányi, András Beke, Steven M. Lulich, ,,Patterns of Hungarian vowel production and perception with regard to subglottal resonances'', The Phonetician, No. 99-100, 2009/2011, pp. 7-28. paper website


2010

·   Csaba Zainkó, Tamás Gábor Csapó, Géza Németh, ,,Special Speech Synthesis for Social Network Websites'', Text, Speech and Dialogue, Lecture Notes in Computer Science, 2010, Vol. 6231/2010, pp. 455-463. paper presentation pic1 pic2

·   Tamás Gábor Csapó, Csaba Zainkó, Géza Németh, ,,A Study of Prosodic Variability Methods in a Corpus-Based Unit Selection Text-To-Speech System'', Infocommunications Journal, Budapest, Vol. LXV. / I., pp. 32-37. abstract paper


2009

·   Tamás Gábor Csapó, Zsuzsanna Bárkányi, Tekla Etelka Gráczi, Tamás Bőhm, Steven M. Lulich, ,,Relation of formants and subglottal resonances in Hungarian vowels'', In Interspeech 2009, Brighton, United Kingdom, pp. 484-487. abstract paper poster pic


2007

·   Géza Németh, Márk Fék, Tamás Gábor Csapó, ,,Increasing Prosodic Variability of Text-To-Speech Synthesizers'', In Interspeech 2007, Antwerp, Belgium, pp. 474-477. abstract paper poster sample sentences pic