Login
 Forum
 
 
Personal data sheet
 Print preview
VALIDITY EXPIRED
personal data approved: 2020. XII. 01.
Personal data
name Péter Mihajlik
name of institution
doctoral school
BME Doctoral School of Informatics (Supervisor)
BME Doctoral School of Electrical Engineering (Academic staff member)
the share of work in the different doctoral schools. BME Doctoral School of Informatics 50%
BME Doctoral School of Electrical Engineering 50%
Contact details
E-mail address mihajliktmit.bme.hu
phone number +36 1 463-3424
Academic title
scientific degree, title Ph.D.
year degree was obtained 2011
discipline to which degree belongs electrical engineering
institution granting the degree BME
Employment
2002 - Budapest University of Technology and Economics
university professor or researcher
Thesis topic supervisor
number of doctoral students supervised until now 1
number of students who fulfilled course requirements 1
students who obtained their degrees:
Balázs Tarján PhD 2021  DSEE

present PhD students:
Idriss Jairi (PhD) (2026/08)  DSI-BME
Mengkedalai Mengkedalai (PhD) (2026/08)  DSI-BME
Yan Meng (PhD) (2026/08)  DSI-BME
  Thesis topic proposals
Research
research area Automatic Speech Recognition, Deep Learning
research field in which current research is conducted electrical engineering
computer sciences
Publications
2023

Linke J., Kádár M.S., Dobsinszki G., Mihajlik P., Kubin G., Schuppler B.: What do self-supervised speech representations encode? An analysis of languages, varieties, speaking styles and speakers, In: Proceedings of the 24th International Speech Communication Association, Interspeech 2023, International Speech Communication Association (ISCA) (2023) pp. 5371-5375.
type of document: Part of book/Proceedings Paper
language: English
URL 
2023

Mihajlik Péter, Kádár Máté Soma, Dobsinszki Gergely, Meng Yan, Kedalai Meng, Linke Julian, Fegyó Tibor, Mády Katalin: What Kind of Multi- or Cross-lingual Pre-training is the most Effective for a Spontaneous, Less-resourced ASR Task?, In: 2nd Annual Meeting of the ELRA/ISCA SIG on Under-resourced Languages (SIGUL 2023), (2023) pp. 58-62.
type of document: Conference paper/Előadás vagy poszter cikke
language: English
URL 
2022

Tarján Balázs, Fegyó Tibor, Mihajlik Péter: Morphology aware data augmentation with neural language models for online hybrid ASR, ACTA LINGUISTICA ACADEMICA 69: (4) pp. 581-598.
type of document: Journal paper/Article
language: English
URL 
2022

Mihajlik Péter, Gráczi Tekla Etelka, Kohári Anna, Tarján Balázs, Balog András, Mády Katalin: A BEA továbbfejlesztése és alkalmazása kontrasztív gépi beszédfelismerési kísérletekre, In: Mády, Katalin; Markó, Alexandra (szerk.) Általános nyelvészeti tanulmányok 34., Akadémiai Kiadó (2022) pp. 361-380.
type of document: Part of book/Szaktanulmány (to be translated)
language: Hungarian
2022

Mihajlik Péter, Balog András, Gráczi Tekla Etelka, Kohári Anna, Tarján Balázs, Mády Katalin: BEA-Base: A Benchmark for ASR of Spontaneous Hungarian, In: Nicoletta, Calzolari; Frédéric, Béchet; Philippe, Blache; Khalid, Choukri; Chritopher, Cieri; Thierry, Declerk; Sara, Goggi; Hitoshi, Isahara; Bente, Maegaard; Joseph, Mariani; Hélene, Mazo; Jan, Odijk; Stelios, Piperidis (szerk.) LREC 2022, Thirteeth International Conference on Language Resources and Evaluation, European Language Resources Association (ELRA) (2022) pp. 1970-1977.
type of document: Part of book/Proceedings Paper
number of independent citations: 1
language: English
Full text 
2021

Mihajlik Péter, Balog András, Tarján Balázs, Fegyó Tibor: End-to-end és hibrid mélyneuronháló alapú gépi leiratozás magyar nyelvű telefonos ügyfélszolgálati beszélgetésekre, In: Berend, Gábor ✉; Gosztolya, Gábor ✉; Vincze, Veronika ✉ (szerk.) XVII. Magyar Számítógépes Nyelvészeti Konferencia : MSZNY 2021, Szegedi Tudományegyetem, Informatikai Intézet (2021) pp. 139-145.
type of document: Part of book/Proceedings Paper
language: Hungarian
2021

Fekete Tímea, Mihajlik Péter: End-to-End Recognition of Spontaneous Speech on the Hungarian BEA Database, BESZÉDTUDOMÁNY / SPEECH SCIENCE 2: (1) pp. 261-272.
type of document: Journal paper/Article
language: English
URL 
2019

Tarján Balázs, Szaszák György, Fegyó Tibor, Mihajlik Péter: Investigation on N-Gram Approximated RNNLMs for Recognition of Morphologically Rich Speech, In: Martín-Vide, Carlos; Purver, Matthew; Pollak, Senja (szerk.) Statistical Language and Speech Processing, Springer Netherlands (2019) pp. 223-234.
type of document: Part of book/Proceedings Paper
number of independent citations: 3
language: English
URL 
2017

Péter Mihajlik, Lili Szabó, Balázs Tarján, András Balog, Krisztina Rábai: First Results in Developing a Medieval Latin Language Charter Dictation System for the East-Central Europe Region, In: Francisco, Lacerda (szerk.) 18th Annual Conference of the International Speech Communication Association, International Speech Communication Association (ISCA) (2017) pp. 2058-2062.
type of document: Part of book/Proceedings Paper
language: English
URL 
2015

Varga A, Tarjan B, Tobler Z, Szaszak G, Fegyo T, Bordas C, Mihajlik P: Automatic Close Captioning for Live Hungarian Television Broadcast Speech: A Fast and Resource-Efficient Approach, LECTURE NOTES IN ARTIFICIAL INTELLIGENCE 9319: pp. 105-112.
type of document: Journal paper/Konferenciaközlemény
number of independent citations: 7
language: English
URL 
Number of independent citations to these publications:11 
Scientometric data
list of publications and citations
number of scientific publications that meet accreditation criteria:
86
number of scientific publications:
84
monographs and professional books:
0
monographs/books in which chapters/sections were contributed:
1 
number of independent citations to scientific publications and creative works:
319


2024. IV. 17.
ODT ülés
Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).

 
All rights reserved © 2007, Hungarian Doctoral Council. Doctoral Council registration number at commissioner for data protection: 02003/0001. Program version: 2.2358 ( 2017. X. 31. )