Scientific Program

Download conference program in pdf format


“SpeD 2013” PROGRAM

October 16, 2013
Technical University of Cluj-Napoca
Aula “Domşa”

9:00 – 9:30

Welcome from the Organizing Committee and Foreword, Corneliu Burileanu, University “Politehnica” of Bucharest, Horia-Nicolai Teodorescu, c.m. of the Romanian Academy, Technical University of Iaşi, Corneliu Rusu, Technical University of Cluj-Napoca.
View this presentation


9:30 – 11:00

Speech Analysis, Representations and Models. Audio Signal Processing

Chairs: Jean-Paul Haton, Corneliu Rusu

  • Invited presentation: Recent Advances in Automatic Speech Recognition,
    Jean-Paul Haton,
    Institut Universitaire de France, LORIA - Université de Lorraine, Nancy.
    View this presentation

  • Invited paper: Singing Voice Identification and Lyrics Transcription for Music Information Retrieval,
    Annamaria Mesaros,
    Department of Signal Processing and Acoustics, Aalto University, Espoo, Finland.

  • Evaluation of Sentiment Polarity Prediction using a Dimensional and a Categorical Approach,
    Ioana Mureşan1, Adriana Stan1, Mircea Giurgiu1, Rodica Potolea2,
    1Department of Communications, Technical University of Cluj-Napoca, Romania,
    2Department of Computer Science, Technical University of Cluj-Napoca, Romania.
    View this presentation

  • Extensive Evaluation Experiments for the Accumulated Cross-Power Spectrum Methods for Time Delay Estimation,
    Radu-Sebastian Marinescu, Andi Buzo, Horia Cucu, Corneliu Burileanu,
    Speech and Dialogue (SpeeD) Laboratory, Faculty of Electronics, Telecommunications and Information Technology, University “Politehnica” of Bucharest, Romania.
    View this presentation

  • Prospective presentation: Voice and Text Analysis as a Tool for Fast Disaster Effects Estimation – A Preliminary Exploration,
    Horia-Nicolai Teodorescu 1,2,
    1Institute for Computer Science, Romanian Academy,
    2Faculty of Electronics, Telecommunications and Information Technology, “Gheorghe Asachi” Technical University of Iaşi, Romania.

11:00 – 11:30    Break and Poster session


11:30 – 12:30

  • Algorithm for Detection of Voice Signal Periodicity,
    Ovidiu Buză1, Gavril Toderean1, József Domokos2,
    1Department of Communications, Technical University of Cluj-Napoca, Romania,
    2Department of Electrical Engineering, Sapientia University, Targu Mures, Romania.
    View this presentation

  • Error Correction Mechanism for Five-Key Chording Keyboards,
    Adrian Tărniceriu, Bixio Rimoldi, Pierre Dillenbourg,
    School of Computer and Communication Sciences, Ecole Polytechnique Fédérale de Lausanne, Switzerland.
    View this presentation

  • A Romanian Language Modeling using Linguistic Factors,
    Marilena Lazăr, Diana Militaru,
    Information Systems and Communications Test & Evaluation Scientific Research Center, Military Equipment and Technologies Research Agency, Bucharest, Romania.
    View this presentation




October 16, 2013
Technical University of Cluj-Napoca
Aula “Domşa”


Spoken Language Recognition. Keyword Spotting and Information Retrieval.
Human-Computer Interfaces

Chairs: Horia-Nicolai Teodorescu, Dragoş Burileanu


14:00 – 16:00

  • Invited paper: On the Design of an Acoustic Based Wildlife Intruder Detection System,
    Corneliu Rusu1, Marius Ghiurcău1, Radu Bîlcu2, Marius Sîrbu3, Jaakko Astola4,
    1Technical University of Cluj-Napoca, Romania,
    2Nokia Research Centre, Tampere, Finland,
    3WsComm Cluj-Napoca, Romania,
    4Tampere International Center for Signal Processing, Finland.
    View this presentation

  • A Comparison of Arabic Speech Recognition for Multi-Dialect vs. Specific Dialects,
    Khalid Almeman, Mark Lee,
    School of Computer Science, University of Birmingham, UK.

  • An Incremental Methodology for Improving Pronunciation Dictionaries for Arabic Speech Recognition,
    Khalid Almeman, Mark Lee,
    School of Computer Science, University of Birmingham, UK.

  • Improved Recognition of Hungarian Call Center Conversations,
    Balázs Tarján1, Gellért Sárosi1, Tibor Fegyó1,2, Péter Mihajlik1,3,
    1Department of Telecommunications and Media Informatics, Budapest University of Technology & Economics, Hungary,
    2AITIA International Inc., Hungary,
    3THINKTech Research Center, Vác, Hungary.
    View this presentation

  • Lightly Supervised Acoustic Model Training for Imprecisely and Asynchronously Transcribed Speech,
    Péter Mihajlik1,2, András Balog1,
    1THINKTech Research Center, Vác, Hungary.
    2Department of Telecommunications and Media Informatics, Budapest University of Technology & Economics, Hungary.
    View this presentation

16:00 – 16:15    Break


16:15 – 17:15

  • Text Spotting in Large Speech Databases for Under-Resourced Languages,
    Andi Buzo, Horia Cucu, Corneliu Burileanu,
    Speech and Dialogue (SpeeD) Laboratory, Faculty of Electronics, Telecommunications and Information Technology, University “Politehnica” of Bucharest, Romania.
    View this presentation

  • Multilingual Query by Example Spoken Term Detection for Under-Resourced Languages,
    Andi Buzo, Horia Cucu, Mihai Safta, Corneliu Burileanu,
    Speech and Dialogue (SpeeD) Laboratory, Faculty of Electronics, Telecommunications and Information Technology, University “Politehnica” of Bucharest, Romania.
    View this presentation

  • Estimating the Hurst Exponent in Motor Imagery-based Brain Computer Interface,
    Roxana Aldea, Daniela Tărniceriu,
    Faculty of Electronics, Telecommunications and Information Technology, “Gheorghe Asachi” Technical University, Iaşi, Romania.
    View this presentation



October 17, 2013
Technical University of Cluj-Napoca
Aula “Domşa”


Natural Language Processing. Speech Translation

Chairs: Dan Tufiş, Dan Cristea


9:00 – 10:40

  • Invited paper: The RACAI Speech Translation System. Challenges of Morphologically Rich Languages,
    Dan Tufiş, Tiberiu Boroş, Ştefan Daniel Dumitrescu,
    RACAI – Research Institute for Artificial Intelligence, Romanian Academy, Bucharest, Romania.
    View this presentation

  • Invited paper: Linking Book Characters. Toward A Corpus Encoding Relations Between Entities,
    Dan Cristea1,2, Eugen Ignat1,
    1“Alexandru Ioan Cuza” University of Iaşi, Department of Computer Science (UAIC-FII),
    2Institute of Computer Science, Romanian Academy, Iaşi, Romania.
    View this presentation

  • A Novel Discriminative Method for Pruning Pronunciation Dictionary Entries,
    Seppo Enarvi, Mikko Kurimo,
    Aalto University School of Electrical Engineering, Department of Signal Processing and Acoustics, Espoo, Finland.
    View this presentation

  • Acknowledgement or Reply? Prosodic Features for Disambiguating Pragmatic Functions of the Italian Token ‘sì’,
    Michelina Savino1, Mario Refice2,
    1Department of Education, Psychology, Communication, University of Bari, Italy,
    2Department of Electrical and Information Engineering, Polytechnics of Bari, Italy.

10:40 – 11:00    Break


11:00 – 12:20

  • An Integrated Language Identification for Code-Switched Speech using Decoded-Phonemes and Support Vector Machine,
    Koena Ronny Mabokela, Madimetja Jonas D. Manamela,
    Telkom Centre of Excellence for Speech Technology (TCoE4ST), Department of Computer Science, University of Limpopo, Turfloop Campus Limpopo Province, Polokwane, South Africa.
    View this presentation

  • Aspects of a Romanian Syntactic-Prosodic Interface for an Intonation Prediction Module,
    Doina Jitcă, Vasile Apopei, Otilia Păduraru,
    Institute of Computer Science of the Romanian Academy, Iaşi branch.
    View this presentation

  • Robust Spectral Representation Using Group Delay Function and Stabilized Weighted Linear Prediction for Additive Noise Degradations,
    Dhananjaya Gowda, Jouni Pohjalainen, Paavo Alku, Mikko Kurimo,
    Aalto University School of Electrical Engineering, Department of Signal Processing and Acoustics, Espoo, Finland.
    View this presentation

  • On Letter to Sound Conversion for Romanian: A Comparison of Five Algorithms,
    Ştefan-Adrian Toma, Traian Bîrsan, Felix Totir, Eugeniu Oancea,
    Military Technical Academy, Bucharest, Romania.
    View this presentation



October 17, 2013
Technical University of Cluj-Napoca
Aula “Domşa”

14:00 – 14:30    Demo Session for the project “Simple4All”


Text-to-Speech Synthesis. Speaker Recognition. Assistive Technologies

Chairs: Inge Gavăt, Gavril Toderean


14:30 – 16:00

  • Statistically Augmented Preprocessing/Normalization Module for a Romanian Text-to-Speech System,
    Cătălin Ungurean, Dragoş Burileanu, Mihai Surmei,
    Speech and Dialogue (SpeeD) Laboratory, Faculty of Electronics, Telecommunications and Information Technology, University “Politehnica” of Bucharest, Romania.
    View this presentation

  • Specific Acoustic Unit Processing in Concatenative Romanian Speech Synthesis Used for Talking Agents,
    Cristian Negrescu, Amelia Ciobanu, Mihai Daniel Ilie,
    Telecommunication Department, Faculty of Electronics, Telecommunications and Information Technology, University “Politehnica” of Bucharest, Romania.

  • Experiments on Using Vocal Tract Estimates of Nasal Stops for Speaker Verification,
    Ewald Enzinger, Christian H. Kasess,
    Acoustics Research Institute, Austrian Academy of Sciences, Vienna, Austria.
    View this presentation

  • On Forensic Speaker Recognition Case Pre-Assessment,
    Gheorghe Pop1, Dragoş Drăghicescu2, Dragoş Burileanu2,
    1National Institute of Forensic Expertise, Bucharest, Romania
    2Speech and Dialogue (SpeeD) Laboratory, Faculty of Electronics, Telecommunications and Information Technology, University “Politehnica” of Bucharest, Romania.
    View this presentation

16:00 – 16:15    Break


16:15 – 17:15

  • Speech Recognition of Aged Voice in the AAL Context: Detection of Distress Sentences,
    Frédéric Aman, Michel Vacher, Solange Rossato, François Portet,
    LIG Laboratory, GETALP Team, UMR CNRS/UJF/INPG/UPMF 5217, Grenoble, France.
    View this presentation

  • Some Aspects of Synthetic Elderly Voices in Ambient Assisted Living Systems,
    Csaba Zainkó, Bálint Tóth, Mátyás Bartalis, Géza Németh, Tibor Fegyó,
    Department of Telecommunications and Media Informatics, Budapest University of Technology and Economics, Hungary.

  • Dantzig Selector for Audio Data Reconstruction,
    Ioan Buciu,
    Department of Electronics and Telecommunications, Faculty of Electrical Engineering and Information Technology, University of Oradea, Romania.
    View this presentation

Concluding remarks, Corneliu Burileanu, University “Politehnica” of Bucharest.


POSTER SESSION
Aula “Domşa” and Hall

Organizers: Constantin Paleologu, Călin Vlădeanu

  • Text-Independent Speaker Verification Using the GMM-UBM Modelling,
    Svetlana Segărceanu, Tiberius Zaharia,
     Department of Applied Electronics and Information Engineering, Faculty of Electronics, Telecommunications and Information Technology, University „Politehnica” of Bucharest, Romania.

  • A Hybrid Approach for Speech Recognition,
    Tiberius Zaharia, Svetlana Segărceanu,
     Department of Applied Electronics and Information Engineering, Faculty of Electronics, Telecommunications and Information Technology, University “Politehnica” of Bucharest, Romania.

  • Real-time Platform for Vocoders,
    András Balogh, Ovidiu Buză, Gavril Toderean,
     Department of Communications, Technical University of Cluj-Napoca, Romania.

  • Unsupervised Speaker Recognition Approach using an Automatic Clustering Algorithm,
    Tudor Barbu,
     Institute of Computer Science of the Romanian Academy, Iasi.