Máté Ákos Tündik

Publication Activity (10 Years)

Years Active: 2016-2020
Publications (10 Years): 16

Top Topics

Segmentation Errors

Spoken Language

Automatic Summarization

Top Venues

Publications

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)
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)
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)
Gábor Járó, Attila Hilt, Loránd Nagy, Máté Ákos Tündik, József Varga
Evolution towards Telco-Cloud: Reflections on Dimensioning, Availability and Operability : (Invited Paper). TSP (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, Attila Hilt, Gergo Bota, Lorand Nagy, Kalle Luukkanen
Access-independent Cloud-based Real-Time Translation Service for Voice Calls in Mobile Networks. CSNDSP (2018)
Máté Ákos Tündik, György Szaszák
Joint Word- and Character-level Embedding CNN-RNN Models for Punctuation Restoration. CogInfoCom (2018)
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)
Máté Ákos Tündik, Balázs Tarján, György Szaszák
Á bilingual comparison of MaxEnt-and RNN-based punctuation restoration in speech transcripts. CogInfoCom (2017)
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)
Máté Ákos Tündik, Branislav Gerazov, Aleksandar Gjoreski, György Szaszák
Atom decomposition based stress detection and automatic phrasing of speech. CogInfoCom (2016)