Dr. Tamás Gábor Csapó

Research Fellow


Budapest University of Technology and Economics
Department of Telecommunications and Media Informatics
BME building I, room IB151
Magyar Tudósok krt. 2.
Budapest, 1117 HUNGARY


csapot /at/ tmit /dot/ bme /dot/ hu



Tamás Gábor Csapó obtained his PhD in computer science & speech technology from Budapest University of Technology and Economics (BME), Hungary in 2014. During MSc, in 2007, he was awarded with 1st prize of the National Conference of Scientific Student’s Associations, Hungary. He received a CIRE student grant of the Acoustical Society of America in 2010 and was a Fulbright scholar at Indiana University, USA in 2014, where he started to deal with ultrasound tongue imaging. In 2016, he joined the MTA-ELTE Lingual Articulation Research Group, focusing on investigating the Hungarian articulation during speech production. Since 2017, he has two national research projects about ultrasound-based articulatory-to-acoustic mapping and articulatory-to-acoustic inversion, both of them applying deep learning methods. He regularly cooperates with international researchers and has co-authors from USA, Canada, Colombia, China, and several EU countries. His research interests include speech synthesis, speech analysis, vocoding, ultrasound-based tongue movement analysis, and deep learning methods applied for speech technologies.

Research areas

Silent Speech Interfaces, deep learning, Speech articulation (2D ultrasound), text-to-speech synthesis, excitation modeling

Programming languages, Frameworks

Programming languages
Matlab/Octave, Python, C#, Xamarin
HTS, hts_engine, SPTK, Merlin
Deep learning
keras, TensorFlow

Language skills



  • Budapest University of Technology and Economics

    Summa cum laude, topic: Increasing the naturalness of synthesized speech in hidden Markov-model based text-to-speech synthesis

  • Department of Speech and Hearing Sciences, Indiana University, Bloomington, IN, USA (Fulbright scholarship)

    Research topic: ,,Investigation of tongue movement during speech using ultrasound''

  • Budapest University of Technology and Economics

    Faculty of Electrical Engineering and Informatics, Technical informatics