György Szaszák

Publication Activity (10 Years)

Years Active: 2004-2023
Publications (10 Years): 35

Top Topics

Air Traffic Control

Automatic Summarization

Speech Recognition

Spoken Language

Top Venues

Publications

Kamel Nebhi, György Szaszák
Automatic Assessment Of Spoken English Proficiency Based on Multimodal and Multitask Transformers. RANLP (2023)
Dávid Sztahó, György Szaszák, András Beke
Deep Learning Methods in Speaker Recognition: A Review. Period. Polytech. Electr. Eng. Comput. Sci. 65 (4) (2021)
Balázs Tarján, György Szaszák, Tibor Fegyó, Péter Mihajlik
Improving Real-time Recognition of Morphologically Rich Speech with Transformer Language Model. CogInfoCom (2020)
Balázs Tarján, György Szaszák, Tibor Fegyó, Péter Mihajlik
On the Effectiveness of Neural Text Generation Based Data Augmentation for Recognition of Morphologically Rich Speech. TDS (2020)
Balázs Tarján, György Szaszák, Tibor Fegyó, Péter Mihajlik
Deep Transformer based Data Augmentation with Subword Units for Morphologically Rich Online ASR. CoRR (2020)
Miklós Gábriel Tulics, György Szaszák, Krisztina Mészáros, Klára Vicsi
Using ASR Posterior Probability and Acoustic Features for Voice Disorder Classification. CogInfoCom (2020)
Balázs Tarján, György Szaszák, Tibor Fegyó, Péter Mihajlik
On the Effectiveness of Neural Text Generation based Data Augmentation for Recognition of Morphologically Rich Speech. CoRR (2020)
Máté Ákos Tündik, Balázs Tarján, György Szaszák
A low latency sequential model and its user-focused evaluation for automatic punctuation of ASR closed captions. Comput. Speech Lang. 63 (2020)
Balázs Tarján, György Szaszák, Tibor Fegyó, Péter Mihajlik
N-gram Approximation of LSTM Recurrent Language Models for Single-pass Recognition of Hungarian Call Center Conversations. CogInfoCom (2019)
Balázs Tarján, György Szaszák, Tibor Fegyó, Péter Mihajlik
Investigation on N-gram Approximated RNNLMs for Recognition of Morphologically Rich Speech. CoRR (2019)
Máté Ákos Tündik, Valér Kaszás, György Szaszák
On the Effects of Automatic Transcription and Segmentation Errors in Hungarian Spoken Language Processing. Period. Polytech. Electr. Eng. Comput. Sci. 63 (4) (2019)
Dávid Sztahó, György Szaszák, András Beke
Deep learning methods in speaker recognition: a review. CoRR (2019)
Máté Ákos Tündik, Valér Kaszás, György Szaszák
Assessing the Semantic Space Bias Caused by ASR Error Propagation and its Effect on Spoken Document Summarization. INTERSPEECH (2019)
Balázs Tarján, György Szaszák, Tibor Fegyó, Péter Mihajlik
Investigation on N-Gram Approximated RNNLMs for Recognition of Morphologically Rich Speech. SLSP (2019)
Miklós Gábriel Tulics, György Szaszák, Krisztina Mészáros, Klára Vicsi
Artificial Neural Network and SVM based Voice Disorder Classification. CogInfoCom (2019)
Bálint Döbrössy, Márton Makrai, Balázs Tarján, György Szaszák
Investigating Sub-Word Embedding Strategies for the Morphologically Rich and Free Phrase-Order Hungarian. RepL4NLP@ACL (2019)
György Szaszák, Máté Ákos Tündik
Leveraging a Character, Word and Prosody Triplet for an ASR Error Robust and Agglutination Friendly Punctuation Approach. INTERSPEECH (2019)
Máté Ákos Tündik, György Szaszák, Gábor Gosztolya, András Beke
User-centric Evaluation of Automatic Punctuation in ASR Closed Captioning. INTERSPEECH (2018)
Valér Kaszás, Máté Ákos Tündik, György Szaszák
A semantic space approach for automatic summarization of documents. CogInfoCom (2018)
György Szaszák, Máté Ákos Tündik, Branislav Gerazov
Prosodic stress detection for fixed stress languages using formal atom decomposition and a statistical hidden Markov hybrid. Speech Commun. 102 (2018)
Máté Ákos Tündik, György Szaszák
Joint Word- and Character-level Embedding CNN-RNN Models for Punctuation Restoration. CogInfoCom (2018)
Anna Moró, György Szaszák
A prosody inspired RNN approach for punctuation of machine produced speech transcripts to improve human readability. CogInfoCom (2017)
Máté Ákos Tündik, Gábor Kiss, David Sztahó, György Szaszák
Assessment of pathological speech prosody based on automatic stress detection and phrasing approaches. CogInfoCom (2017)
Máté Ákos Tündik, Balázs Tarján, György Szaszák
Low Latency MaxEnt- and RNN-Based Word Sequence Models for Punctuation Restoration of Closed Caption Data. SLSP (2017)
Youssef Oualil, Dietrich Klakow, György Szaszák, Ajay Srinivasamurthy, Hartmut Helmke, Petr Motlícek
A context-aware speech recognition and understanding system for air traffic control domain. ASRU (2017)
Máté Ákos Tündik, Balázs Tarján, György Szaszák
Á bilingual comparison of MaxEnt-and RNN-based punctuation restoration in speech transcripts. CogInfoCom (2017)
Ajay Srinivasamurthy, Petr Motlícek, Ivan Himawan, György Szaszák, Youssef Oualil, Hartmut Helmke
Semi-Supervised Learning with Semantic Knowledge Extraction for Improved Speech Recognition in Air Traffic Control. INTERSPEECH (2017)
Anna Moró, György Szaszák
A Phonological Phrase Sequence Modelling Approach for Resource Efficient and Robust Real-Time Punctuation Recovery. INTERSPEECH (2017)
Bálint Pál Tóth, Kornél István Kis, György Szaszák, Géza Németh
Ensemble Deep Neural Network Based Waveform-Driven Stress Model for Speech Synthesis. SPECOM (2016)
András Beke, György Szaszák
Automatic Summarization of Highly Spontaneous Speech. SPECOM (2016)
Milan Secujski, Branislav Gerazov, Tamás Gábor Csapó, Vlado Delic, Philip N. Garner, Aleksandar Gjoreski, David Guennec, Zoran A. Ivanovski, Aleksandar Melov, Géza Németh, Ana Stojkovic, György Szaszák
Design of a Speech Corpus for Research on Cross-Lingual Prosody Transfer. SPECOM (2016)
György Szaszák, Máté Ákos Tündik, Branislav Gerazov, Aleksandar Gjoreski
Combining Atom Decomposition of the F0 Track and HMM-based Phonological Phrase Modelling for Robust Stress Detection in Speech. SPECOM (2016)
György Szaszák, Máté Ákos Tündik, András Beke
Summarization of Spontaneous Speech using Automatic Speech Recognition and a Speech Prosody based Tokenizer. KDIR (2016)
Gábor Gosztolya, Tamás Grósz, György Szaszák, László Tóth
Estimating the Sincerity of Apologies in Speech by DNN Rank Learning and Prosodic Analysis. INTERSPEECH (2016)
Máté Ákos Tündik, Branislav Gerazov, Aleksandar Gjoreski, György Szaszák
Atom decomposition based stress detection and automatic phrasing of speech. CogInfoCom (2016)
György Szaszák, András Beke, Gábor Olaszy, Bálint Pál Tóth
Using automatic stress extraction from audio for improved prosody modelling in speech synthesis. INTERSPEECH (2015)
Ádam Varga, Balázs Tarján, Zoltán Tobler, György Szaszák, Tibor Fegyó, Csaba Bordás, Péter Mihajlik
Automatic Close Captioning for Live Hungarian Television Broadcast Speech: A Fast and Resource-Efficient Approach. SPECOM (2015)
György Szaszák, András Beke
Toward Exploring the Role of Disfluencies from an Acoustic Point of View: A New Aspect of (Dis)continuous Speech Prosody Modelling. TSD (2015)
András Beke, György Szaszák
Combining NLP techniques and acoustic analysis for semantic focus detection in speech. CogInfoCom (2014)
András Beke, György Szaszák, Viola Varadi
Automatic phrase segmentation and clustering in spontaneous speech. CogInfoCom (2013)
György Szaszák, András Beke
Using phonological phrase segmentation to improve automatic keyword spotting for the highly agglutinating Hungarian language. INTERSPEECH (2013)
György Szaszák, Philip N. Garner
Evaluating intra- and crosslingual adaptation for non-native speech recognition in a bilingual environment. CogInfoCom (2013)
András Beke, György Szaszák
Unsupervised Clustering of Prosodic Patterns in Spontaneous Speech. TSD (2012)
György Szaszák, András Beke
Automatic prosodic and syntactic analysis from speech in cognitive infocommunication. CogInfoCom (2012)
György Szaszák, András Beke
Exploiting Prosody for Syntactic Analysis in Automatic Speech Understanding. J. Lang. Model. (1) (2012)
György Szaszák, Katalin Nagy, András Beke
Analysing the Correspondence Between Automatic Prosodic Segmentation and Syntactic Structure. INTERSPEECH (2011)
Fabien Ringeval, Jean Demouy, György Szaszák, Mohamed Chetouani, L. Robel, Jean Xavier, David Cohen, Monique Plaza
Automatic Intonation Recognition for the Prosodic Assessment of Language-Impaired Children. IEEE Trans. Speech Audio Process. 19 (5) (2011)
Klára Vicsi, György Szaszák
Using prosody to improve automatic speech recognition. Speech Commun. 52 (5) (2010)
György Szaszák, David Sztahó, Klára Vicsi
Automatic intonation classification for speech training systems. INTERSPEECH (2009)
Klára Vicsi, György Szaszák
Using prosody for the improvement of ASR - sentence modality recognition. INTERSPEECH (2008)
György Szaszák, Klára Vicsi
Speech Recognition Supported by Prosodic Information for Fixed Stress Languages. TSD (2007)
György Szaszák, Klára Vicsi
Using Prosody in Fixed Stress Languages for Improvement of Speech Recognition. COST 2102 Workshop (Vietri) (2007)
Klára Vicsi, György Szaszák
Prosodic Cues for Automatic Phrase Boundary Detection in ASR. TSD (2006)
Klára Vicsi, György Szaszák
Automatic Segmentation of Continuous Speech on Word Level Based on Supra-segmental Features. Int. J. Speech Technol. 8 (4) (2005)
Andrej Zgank, Zdravko Kacic, Frank Diehl, Klára Vicsi, György Szaszák, Jozef Juhár, Slavomír Lihan
The COST 278 MASPER Initiative - Crosslingual Speech Recognition with Large Telephone Databases. LREC (2004)
György Szaszák, Klára Vicsi
Examination of Pronunciation Variation from Hand-Labelled Corpora. TSD (2004)