Lecturer, PhD, Faculty of Electronics, Telecommunications and Information Technology,
University Politehnica of Bucharest

Email: horia.cucu@upb.ro
Tel: +4021 402 4635
Office: B042, Leu Campus

---

Degrees and Professional Experience Degrees Professional experience Research Interests Academic Activity Teaching Teaching books Strategic programmes National collaborations PhD student supervision Bachelor and Master student supervision Last update: July 2014

Scientific Activity Scientific traceroute and achievements International research projects National research projects International Collaborations National Collaborations Benchmarking Affiliation to scientific community List of publications PhD Thesis Books and book chapters Journal and conference papers List of citations

---

Research Interests

• Natural Language Processing (Statistical Language Modeling, Phonetization, Diacritics restoration)
• Spoken Language Technology (Automatic Speech/Speaker Recognition, Speech Indexing, Spoken Term Detection)
• Statistical Machine Translation

Degrees and Professional Experience

Degrees

• PhD Degree, in Electronics and Telecommunications (Oct 2011), University “Politehnica” of Bucharest.
• Engineer Degree, in Applied Electronics (June 2008), University “Politehnica” of Bucharest.
• Bachelor Degree (June 2003), “Dr. Ioan Mesota” National College, Braşov.

Professional experience

• Lecturer (Oct 2012 – present) at Faculty of Electronics, Telecommunications and Information Technology, University Politehnica of Bucharest. Teach Microcontrollers and Embedded Systems, Microprocessor Architectures and Spoken Language Technology. Supervised BSc, MSc and PhD projects in speech recognition and embedded systems.
• Research Engineer (Jan 2014 – present) at University Politehnica of Bucharest, in the eWALL project (research and development project funded by the European Commission through the 7th Framework Programme).
• Project Manager and Research Engineer (Oct 2013 – Jun 2014) at University Politehnica of Bucharest, in the ​LVCSR-ROM p​roject (research and development project funded by the Romanian-American Foundation, Bucharest, Romania). Coordinated a team of four senior researchers. Developed several enhancement modules for the Large Vocabulary Continuous Speech Recognition (LVCSR) system for the Romanian language. Created a web-service which provides rich speech transcriptions for multimedia files.
• IT Consultant (Dec 2013 – Jan 2014) for Intelligent IT, Sibiu, Romania. Designed and implemented a Natural Language Processing software module for a personal assistant smartphone application (OmniBuddy).
• Implementing Expert (Jun 2013 – Sep 2013) at University Politehnica of Bucharest, in the CASIA project (FSE – European Structural Funds POS-DRU project), owner Research Institute for Artificial Intelligence, Romanian Academy. Coordinated and supervised 14 students for their summer internships in speech and language processing.
• Implementing Expert (Jan 2011 – Aug 2013) at University Politehnica of Bucharest, in the PROMISE project (FSE – European Structural Funds POS-DRU project), owner University Politehnica of Bucharest. Developed the teaching infrastructure for several subjects within a new Master programme (BIOSINF).
• Teaching Assistant (Oct 2008 – Oct 2012) at Faculty of Electronics, Telecommunications and Information Technology, University Politehnica of Bucharest. Teaching Microcontrollers and Embedded Systems, Microprocessor Architectures, Object-Oriented Programming and Spoken Language Technology. Supervising BSc and MSc projects in speech recognition and embedded systems.
• IT Consultant (Jan 2011 – Feb 2011) for RSM Scot, Bucharest, Romania. Designed and implemented an employees timesheet software application.
• IT Consultant (Oct 2009 – June 2010) for Grob Technologies Inc., Massachusetts, USA. Designed and implemented the server-side software application of a web-based service for social-networking tracking (What You Post).
• Research Engineer (Aug 2007 – Mar 2009) at University Politehnica of Bucharest, in the PALIROM project (research and development project funded by the Romanian Government, owner Softwin Group, Bucharest, Romania). Developed natural language resources for Romanian, implemented and tested a natural language compiler.
• Research Engineer (Aug 2007 – Mar 2009) at University Politehnica of Bucharest, in the BIOACS project (research and development project funded by the Romanian Government, owner Softwin Group, Bucharest, Romania). Collected a database of digital handwritten signatures and developed digital signal processing algorithms for handwritten signature recognition.
• Software Engineer (Jan 2006 – Jan 2009) at Ubicore Technology, Bucharest, Romania. Designed and implemented several video processing algorithms on a new massive parallel CPU, developed a distributed-computing application for highly computational tasks, developed a complete debugger tool for a new CPU architecture.

Academic Activity

Teaching

All my teaching activity takes place in the Faculty of Electronics, Telecommunications and Information Technology, University “Politehnica” of Bucharest.

Master curricula

• Spoken Language Technology, research project (2012 – present);
• Microcontrollers and Embedded Systems, laboratory (2010 – present).

Bachelor curricula

• Microprocessors Architecture, course (2012 – present) and laboratory (2008 – present);
• Microcontrollers, course (2012 – present) and laboratory (2008 – present);
• Object Oriented Programming (Java), laboratory (2010).

Teaching books

• Electronic support for “Microcontrollers” laboratory;
• Electronic support for “Spoken Language Technology” research project.

Strategic programmes

• 2013: implementing expert, project CASIA (“Support for a successful career in artificial intelligence”), FSE – European Structural Funds POS-DRU project, owner University „Politehnica” of Bucharest, ID POSDRU/109/2.1/G/81772
• 2010-2013: implementing expert, project PROMISE (“An Integrated Master’s Degree Programme in the Fields of Sound, Image and Multimedia Engineering”), FSE – European Structural Funds POS-DRU project, owner University „Politehnica” of Bucharest, ID61178.

National collaborations

• Romanian Association for Artificial Intelligence (ARIA): “HMMs: from Theory to Applications”, a 2-day hands-on workshop on some of the most well-known machine learning models used in symbol and speech recognition.

PhD student supervision

Since 2012, I was part of the PhD Supervision Commission for several PhD students:

• Mircea Sorin Rusu: “”, University Politehnica of Bucharest (supervisor prof. C. Burileanu).
• Gheorghe Pop: “”, University Politehnica of Bucharest (supervisor prof. C. Burileanu).
• Aurelian Derviş: “Contribuţii la sinteza vorbirii în limba română”, University Politehnica of Bucharest (supervisor prof. C. Burileanu).
• Andrei Purică: “Semantic Video Coding”, co-tutelle thesis between University Politehnica of Bucharest (supervisor prof. C. Burileanu) and Telecom ParisTech (supervisor prof. F. Dufaux, prof. B. Pesquet).
• Anca-Livia Radu, “Large Scale Media Analysis via Media Fusion and Crowdsourcing”, co-tutelle thesis between University Politehnica of Bucharest (supervisor prof. C. Burileanu) and University of Trento (supervisor prof. F. Giunchiglia).

Bachelor and Master student supervision

Since 2013, I supervised and coordinated over 20 students during the SpeeD “Speech&Language” Internships.

Since 2009, I co-supervised/coordinated numerous Bachelor of Science and Master of Science projects in Spoken Language Technology, Object Oriented Programming and Embedded Systems:

2014

• Ancuţa Ardeleanu, “Speech-based gender detection and multilingual spoken term detection“.
• Ioana Calangiu, “Multilingual automatic speech recognition system“.
• Mihai Dogariu, “Language identification system“.
• Diana Enescu, “Android telephone calls transcription and searchable history“.
• Radu Harapu, “Speaker recognition web-service“.
• Călin Necula, “Speech-based caller identification. Application for call centers“.
• Alexandru Şerban, “Radio broadcast news transcription web-service“.

2013

• Mihai Cătălin Safta, “Spoken Term Detection for Romanian Language”, SpeeD, University Politehnica of Bucharest, Romania.

2012

• Sorin Duţulescu, “Management application for Internet Service Providers (ISP)”, University Politehnica of Bucharest, Romania.
• Alexandra Jica, “Topic-Based Language Model Adaptation for Automatic Speech Recognition”, SpeeD, University Politehnica of Bucharest, Romania.
• Adrian Liţă, “Multi-Processor 6-Propeller Tridimensional Flying Apparatus”, University Politehnica of Bucharest, Romania.
• Florin Matei, “Home Automation System Using Speech Recognition on an Embedded Platform”, SpeeD, University Politehnica of Bucharest, Romania.
• Cătălin Stănculescu, “Continuous Speech Recognition Agent for Mobile Devices”, SpeeD, University Politehnica of Bucharest, Romania.
• Iulia Stănoiu, “Noisy Speech Enhancement in Automatic Speech Recognition for Romanian Language”, SpeeD, University Politehnica of Bucharest, Romania.
• Vanessa Voinic, “Continuous Speech Recognition in Romanian Language for Medical Applications”, SpeeD, University Politehnica of Bucharest, Romania.

2011

• Daria Ion, “Speaker-dependent speech recognition: An in-depth analysis of speech features”, SpeeD, University Politehnica of Bucharest, Romania.
• Aniela Milea, “Time-domain analysis and compression methods applied to the speech signal”, SpeeD, University Politehnica of Bucharest, Romania.
• Radu-Mihai Pană-Tălpeanu, “Language models in speaker-dependent speech recognition”, SpeeD, University Politehnica of Bucharest, Romania.

2010

• Tudor Mihailescu, Ioana Rolea, “Speaker recognition systems”, SpeeD, University Politehnica of Bucharest, Romania.
• Adina Popa, Diana Uzum, “Speaker-dependent speech recognition”, SpeeD, University Politehnica of Bucharest, Romania.

Scientific Activity

Scientific research trace-route and achievements

• Automatic speech recognition for Romanian (2008 – 2011). Key role in a team which developed the first Large Vocabulary Continuous Speech Recognition (LVCSR) system for Romanian in 2011. Developed the first statistical language model for Romanian in 2011.
• Digital signal processing (2007 – 2009). Part of a team which developed several dynamic (accelerometer-based) handwritten signature recognition techniques.
• Natural language processing (2007 – 2009). Part of a team which developed a natural language compiler for Romanian and several other natural language applications for Romanian.
• Image and video processing (2006 – 2007). Ported state-of-the-art scaling, de-noising and de-interlacing algorithms on a new massive parallel CPU.

International research projects

• 2014 – 2016: research engineer (in the UPB research team), project eWALL (“eWALL for Active Long Living”), funded by the European Commission through the 7th Framework Programme, ID FP7-ICT-2013-10, no. 610658.

National research projects

• 2013-2014: project manager, project LVCSR-ROM (“Noise-robust, domain-adaptable, large-vocabulary automatic speech recognition system for the Romanian language”), funded by Romanian-American Foundation through the Applied Research, Technological Innovation and Entrepreneurship (ARTIE) Fellowship Program.
• 2007-2009: research engineer, project PALIROM (“Applications Package for the Romanian Language”), funded by the Romanian Government through the National Research Authority (“Inovare” programme), owner Softwin Group, ID 10018/26.09.2007.
• 2007-2009: research engineer, project BIOACS (“Biometric System for the Acquisition and Verification of Dynamic Signature”), funded by the Romanian Government through the National Research Authority (“Inovare” programme), owner Softwin Group, ID 10143/28.09.2007.

Patents

• Andi Buzo, Horia Cucu, Lucian Petrică and Dragoş Burileanu, “Metodă și sistem pentru diarizare în timp real a semnalelor audio, utilizate pentru recunoașterea automată a vorbirii și a vorbitorului” (Method and system for real-time diarization of audio signals, with applications in automatic speech and speaker recognition), patent application registered at OSIM, no. A2014/00346.
• Lucian Petrică, Horia Cucu and Andi Buzo, “Metodă pentru restaurarea automată a semnelor diacritice, folosind texte achiziționate electronic, utilizată în procesarea limbajului natural” (Automatic diacritics restoration method using electronically collected texts with applications in natural language processing), patent application registered at OSIM, no. A2014/00347.

International collaborations

• Laboratoire d’Informatique de Grenoble (LIG), France: PhD Internship (Feb 2011 – June 2011), under the supervision of prof. Laurent Besacier.

National collaborations

• Softwin R&D: several research projects in digital signal processing and natural language processing.

Benchmarking

• MediaEval 2013 (Benchmarking Initiative for Multimedia Evaluation): Spoken Web Search Task with LAPI & SpeeD, University Politehnica of Bucharest, Romania.
• MediaEval 2012 (Benchmarking Initiative for Multimedia Evaluation): Spoken Web Search Task with LAPI & SpeeD, University Politehnica of Bucharest, Romania.

Affiliation to scientific community

• Member of EURASIP, Jan 2013 – present;
• Member of SpeeD Laboratory, Oct 2008 – present;
• Peer reviewer for Springer’s Language Resources and Evaluation Journal, Oct 2013 – present;
• Peer reviewer for UPB Scientific Bulletin (Series C), Jan 2013 – present;
• Peer reviewer for SpeD2011, EUSIPCO2012, SpeD2013.

List of publications

PhD thesis

Horia Cucu, “Towards a speaker-independent, large-vocabulary continuous speech recognition system for Romanian”, PhD Thesis, University “Politehnica” of Bucharest, Oct 2011 (scientific coordinator: prof. Corneliu Burileanu).

Books and book chapters

• Corneliu Burileanu, Cristina Sorina Petrea, Andi Buzo, and Horia Cucu, “Speech Recognition Experiments Starting from Isolated Words for Spoken Romanian Language”, book chapter in D. Tufiş, Corina Forăscu (Eds.), “Multilinguality and Interoperability in Language Processing with Emphasis on Romanian”, Publishing House of the Romanian Academy, Bucharest 2010, pp. 229-242, ISBN: 978-973-27-1972-5.
• Corneliu Burileanu, Cristina-Sorina Petrea, Andi Buzo, Horia Cucu and Alina Pasca, “Report on building a tool for Romanian spontaneous speech recognition”, in “The Phonetician”, Number 97 / 2008-I-II, 2008, pp. 68-98, ISSN: 0741-6164.

Journal papers

• Horia Cucu, Andi Buzo, Laurent Besacier, Corneliu Burileanu, “SMT-based ASR Domain Adaptation Methods for Under-Resourced Languages: Application to Romanian“, in Speech Communication Journal, Vol. 56 – Special Issue on Processing Under-Resourced Languages, pp. 195-212, 2014, ISSN: 0167-6393.
• Radu-Sebastian Marinescu, Andi Buzo, Horia Cucu, Corneliu Burileanu, “Applying the Accumulation of Cross-Power Spectrum Technique for Traditional Generalized Cross-Correlation Time Delay Estimation”, in International Journal on Advances in Telecommunications, vol 6, no 3-4, Dec 2013, pp 98-108, ISSN: 1942-2601.
• Horia Cucu, Andi Buzo, Corneliu Burileanu, “Optimization methods for large vocabulary, isolated words recognition in Romanian language”, University “Politehnica” of Bucharest Scientific Bulletin, Series C, vol. 73, issue 2, Bucharest, 2011, pp. 179-192, ISSN: 1454-234x.

Conference papers

2014

• Horia Cucu, Andi Buzo, Lucian Petrică, Dragoş Burileanu and Corneliu Burileanu, “Recent Improvements of the SpeeD Romanian LVCSR System“, in the Proceedings of the 10th International Conference on Communications (COMM), Bucharest, 2014, pp. 111-114.
• Andi Buzo, Horia Cucu, Lucian Petrică and Dragoş Burileanu, “An Automatic Speech Recognition Solution with Speaker Identification Support“, in the Proceedings of the 10th International Conference on Communications (COMM), Bucharest, 2014, pp. 119-122.
• Lucian Petrică, Horia Cucu, Andi Buzo, and Corneliu Burileanu, “A robust diacritics restoration system using unreliable raw text data“, in the Proceedings of the 4th International Workshop on Spoken Language Technologies for Under-resourced Languages (SLTU), St. Petersburg, Russia, 2014, pp. 215-221.
• Horia Cucu, Andi Buzo and Corneliu Burileanu, “Unsupervised acoustic model training using multiple seed ASR systems“, in the Proceedings of the 4th International Workshop on Spoken Language Technologies for Under-resourced Languages (SLTU), St. Petersburg, Russia, 2014, pp. 124-130.
• Valentin Andrei, Corneliu Burileanu, Horia Cucu, Andi Buzo, “Detecting the number of competing speakers – human selective hearing versus spectrogram distance based estimator”, in the Proceedings of the 15th Annual Conference of the International Speech Communication Association (Interspeech), Singapore, 2014, pp. 467-470.

2013

• Andi Buzo, Horia Cucu, Corneliu Burileanu, “Text Spotting In Large Speech Databases For Under-Resourced Languages”, in the Proceedings of the 7th Conference on Speech Technology and Human-Computer Dialogue (SpeD), Cluj-Napoca, 2013, pp. 77-82, ISBN: 978-1-4799-1065-6.
• Andi Buzo, Horia Cucu, Mihai Safta, Corneliu Burileanu, “Multilingual Query by Example Spoken Term Detection for Under-Resourced Languages”, in the Proceedings of the 7th Conference on Speech Technology and Human-Computer Dialogue (SpeD), Cluj-Napoca, 2013, pp. 83-88, ISBN: 978-1-4799-1065-6.
• Radu-Sebastian Marinescu, Andi Buzo, Horia Cucu, Corneliu Burileanu, “Extensive Evaluation Experiments for the Accumulated Cross-Power Spectrum Methods for Time Delay Estimation”, in the Proceedings of the 7th Conference on Speech Technology and Human-Computer Dialogue (SpeD), Cluj-Napoca, 2013, pp. 29-34, ISBN: 978-1-4799-1065-6.
• Andi Buzo, Horia Cucu, Iris Molnar, Bogdan Ionescu and Corneliu Burileanu, “SpeeD @ MediaEval 2013: A Phone Recognition Approach to Spoken Term Detection”, in Working Notes Proceedings of the MediaEval 2013 Workshop, Barcelona, Spain, 2013, ISSN: 1613-0073.
• Horia Cucu, Andi Buzo, Laurent Besacier, Corneliu Burileanu, “Statistical Error Correction Methods for Domain-Specific ASR Systems”, in A.-H. Dediu et al. (Eds.): “Statistical Language and Speech Processing 2013”, LNAI, vol. 7978, Springer-Verlag Berlin Heidelberg, Tarragonna, Spain, 2013, pp. 83–92.
• Radu-Sebastian Marinescu, Andi Buzo, Horia Cucu, Corneliu Burileanu, “Fast Accurate Time Delay Estimation Based on Enhanced Accumulated Cross-Power Spectrum Phase”, in the Proceedings of the 20th European Signal Processing Conference (EUSIPCO), Marrakech, Morocco, 2013.
• Andi Buzo, Horia Cucu, Corneliu Burileanu, “Investigating Image Processing Based Aligner for Large Texts”, in the Proceedings of the 8th International Conference on Digital Telecommunications (ICDT), Venice, Italy, 2013, pp. 50-54, ISBN: 978-1-61208-262-2.
• Radu-Sebastian Marinescu, Andi Buzo, Horia Cucu, Corneliu Burileanu, “New Considerations for Accumulated ρ-Cross Power Spectrum Phase with Coherence Time Delay Estimation”, in the Proceedings of the 8th International Conference on Digital Telecommunications (ICDT), Venice, Italy, 2013, pp. 55-59, ISBN: 978-1-61208-262-2.

2012

• Horia Cucu, Laurent Besacier, Corneliu Burileanu, Andi Buzo, “ASR Domain Adaptation Methods for Low-Resourced Languages: Application to Romanian Language”, in the Proceedings of the 20th European Signal Processing Conference (EUSIPCO), Bucharest, Romania, 2012, pp. 1648-1652, ISSN: 2076-1465.
• Miruna Stănescu, Horia Cucu, Andi Buzo, Corneliu Burileanu, “ASR for Low-Resourced Languages: Building a Phonetically Balanced Romanian Speech Corpus”, in the Proceedings of the 20th European Signal Processing Conference (EUSIPCO), Bucharest, Romania, 2012, pp. 2060-2064, ISSN: 2076-1465.
• Miruna Stănescu, Andi Buzo, Horia Cucu, Corneliu Burileanu, “Statistical Phonetic Analysis of the Romanian Language for Speech Recognition and Synthesis Tasks”, in the Proceedings of 54th International Symposium “ELMAR 2012”, Zadar, Croatia, 2012, pp. 219-222, ISBN: 978-953-7044-13-8.
• A. Buzo, H. Cucu, M. Safta, B. Ionescu and C. Burileanu, “ARF @ MediaEval 2012: A Romanian ASR-based Approach to Spoken Term Detection”, in Working Notes Proceedings of the MediaEval 2012 Workshop, Pisa, Italy, 2012, ISSN: 1613-0073.
• Bogdan Ionescu, Ionut Mironica, Klaus Seyerlehner, Peter Knees, Jan Schlüter, Markus Schedl, Horia Cucu, Andi Buzo, and Patrick Lambert, “ARF @ MediaEval 2012: Multimodal Video Classification”, in Working Notes Proceedings of the MediaEval 2012 Workshop, Pisa, Italy, 2012, ISSN: 1613-0073.

2011

• Horia Cucu, Laurent Besacier, Corneliu Burileanu, Andi Buzo, “Enhancing Automatic Speech Recognition for Romanian by Using Machine Translated and Web-based Text Corpora”, in the Proceedings of The 14th International Conference “Speech and Computer” (SPECOM), Kazan, Russia, 2011, pp. 81-88, ISBN: 978-5-88983-395-6.
• Horia Cucu, Laurent Besacier, Corneliu Burileanu, Andi Buzo, “Investigating the Role of Machine Translated Text in ASR Domain Adaptation: Unsupervised and Semi-supervised Methods”, IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), 2011, Hawaii, USA, pp. 260-265, ISBN: 978-1-4673-0366-8.
• Andi Buzo, Horia Cucu, Corneliu Burileanu, “Improving Automatic Speech Recognition Robustness for the Romanian Language”, in the Proceedings of the 19th European Signal Processing Conference (EUSIPCO), Barcelona, Spain, 2011, pp. 2119-2122, ISSN 2076-1465.
• Andi Buzo, Horia Cucu, Corneliu Burileanu, Miruna Paşca, Vladimir Popescu, “Word error rate improvement and complexity reduction in automatic speech recognition by analyzing acoustic model uncertainty and confusion”, in the Proceedings of the 6th Conference on Speech Technology and Human-Computer Dialogue (SpeD), Braşov, 2011, pp. 67-74, ISBN 978-1-4577-0439-0.

2010

• Corneliu Burileanu, Andi Buzo, Cristina Sorina Petre, Diana Ghelmez-Hanes, Horia Cucu, “Romanian spoken language resources and annotation for speaker independent spontaneous speech recognition”, in the Proceedings of the 5th International Conference on Digital Telecommunications (ICDT), Athens, Greece, 2010, pp. 7-10.
• Cristina Petrea, Andi Buzo, Horia Cucu, Miruna Pasca, Corneliu Burileanu, “Speech Recognition Experimental Results for Romanian Language”, in the Proceedings of The 6th European Conference on Intelligent Systems and Technologies (ECIT), Iaşi, Romania, 2010, Invited Plenary Session I.

List of citations

A list of citations for my research work is provided by the Google Scholar Citations web-service.