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

Email: lucian.georgescu@speed.pub.ro

Research Interests

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

Education and Professional Experience

Education

• 07.2018 – present: PhD Student
  • Faculty of Electronics, Telecommunications and Information Technology
    University Politehnica of Bucharest, Romania
  • Thesis: Methods and technologies of artificial intelligence applied in speech technology
• 10.2016 – 06.2018:  Master of Science
  • Faculty of Electronics, Telecommunications and Information Technology
    University Politehnica of Bucharest, Romania
  • Specialty: Multimedia Technologies in Biometrics and Information Security Applications (BIOSINF)
• 10.2011 – 09.2016:  Bachelor of Science
  • Faculty of Electronics, Telecommunications and Information Technology
    University Politehnica of Bucharest, Romania
  • Specialty: Computers and Information Technology (CTI)
• 07.2014 – 09.2014:  Certificate Linux 1 – Fundamentals
  • Telecom Academy, Bucharest, Romania
• 10.2011 – 06.2014:  Psycho – Pedagogical Module
  • Teacher Training Department, University Politehnica of Bucharest, Romania
• 09.2007 – 06.2011:  Baccalaureate diploma
  • St. Sava National College, Bucharest, Romania
  • Specialty: Mathematics – Informatics
• 05.2011:  Certificate of professional competence (Informatics)
  • St. Sava National College, Bucharest, Romania

Professional experience

• Oct 2016 – present: Research Engineer
  •  Speech & Dialogue Laboratory (SpeeD), University Politehnica of Bucharest, Romania
• 01.2019 – 07.2019: Research Intern
  • Xilinx Research Labs, Dublin (Ireland)
  • Automatic Speech Recognition experiments using various state-of-the-art neural network based algorithms
  • In-depth profiling in terms of model size, number of operations, number of activations
  • Dataflow simulation in order to extract some parallelism and parametrization info
  • Survey paper on deep learning architectures for automatic speech recognition
• 10.2015 – 12.2015: Intern
  • Ixia, Bucharest (Romania)
  • Development of a keyword spotting application
• 01.2015 – 03.2015: Intern
  • Ixia, Bucharest (Romania)
  • Development of an automatic speech recognition system with reduced vocabulary for English language
• 06.2014 – 08.2014: Summer Intern
  • Speech & Dialogue Laboratory (SpeeD), University Politehnica of Bucharest, Romania
  • Development of automatic speech recognition systems
  • Development of web service systems

Research & Development Activity

Scientific research trace-route and achievements

• Automatic speech recognition for Romanian language (2015 – 2017). Key role in a team which improved the first Large Vocabulary Continuous Speech Recognition (LVCSR) system for Romanian. Added acoustic modeling based on deep neural networks (DNN). More info: https://speed.pub.ro/live-transcriber-2017/
• Speaker recognition (2014 – 2015). Key role in a team which developed a voice login web service based on speaker verification and speech recognition. More info: https://speed.pub.ro/voice-login

National research projects

• Research assistant (2018 – present) at University Politehnica of Bucharest, in the ReTeRom project, TADARAV sub-project (research and development project funded by the Romanian Government through UEFISCDI, project coordinator Research Institute for Artificial Intelligence, Romanian Academy).
• Research assistant (2017 – present) at University Politehnica of Bucharest, in the SPIA-VA project (research and development project funded by the Romanian Government through UEFISCDI, project coordinator University Politehnica of Bucharest, Bucharest, Romania).
• 2016 – 2017: participated as research engineer in  project ANVSIB (“Natural-language, Voice-controlled Assistive System for Intelligent Buildings”), funded by the Romanian Government through UEFISCDI, project coordinator University Politehnica of Bucharest, ID PN-II-PT-PCCA-2013-4-0789.
• 2016 – 2017: participated as research engineer in project SPLANN (“Automatic Baby-Language Recognition System”), funded by the Romanian Government through UEFISCDI, project coordinator Softwin Group, ID PN-II-PT-PCCA-2013-4-1443.

Awards and Participations

• Prize of the category at the innovation contest in the field of national security, PatriotFest: “Speech Transcriber” (https://transcriptions.speed.pub.ro), Agility of Knowledge section, organized by the Ministry of National Defence, the Ministry of Internal Affairs, the Romanian Intelligence Service, Special Communications Service, and the Protocol and Guard Service, in partnership with the Association New Strategy Center, Bucharest 2019.
• Second prize at the innovation contest in the field of national security, PatriotFest: “Live Transcriber” (https://speed.pub.ro/live-transcriber-2017), Agility of Knowledge section, organized by the Ministry of National Defence, the Ministry of Internal Affairs, the Romanian Intelligence Service, Special Communications Service, and the Protocol and Guard Service, in partnership with the Association New Strategy Center, Bucharest, 2018.
• Participation at the Session of Scientific Papers: “Autommatic annotation of speech corpora using complementary GMM and DNN acoustic models”, organized by the Electronic Devices, Circuits and Architectures (DCAE) department from Faculty of Electronics, Telecommunications and Information Technology, University Politehnica of Bucharest, 2018.
• Second prize at the Session of Scientific Papers: “Speaker recongnition based on GMM-UBM models using a Romanian speech corpora”, organized by the Electronic Devices, Circuits and Architectures (DCAE) department from Faculty of Electronics, Telecommunications and Information Technology, University Politehnica of Bucharest, 2018.
• Second prize at the innovation contest in the field of national security, PatriotFest: “Web authentication service based on vocal mark”, Human Machine Interface section, organized by the Ministry of National Defence, the Ministry of Internal Affairs, the Romanian Intelligence Service, Special Communications Service, and the Protocol and Guard Service, in partnership with the Association New Strategy Center, Bucharest, 2017.
• Third prize at the Session of Scientific Papers: “SpeeD’s DNN approach to Romanian speech recognition”, organized by the Electronic Devices, Circuits and Architectures (DCAE) department from Faculty of Electronics, Telecommunications and Information Technology, University Politehnica of Bucharest, 2016.
• Second prize at the Session of Scientific Papers: “Web authentication service based on vocal mark”, organized by the Telecommunications department from Faculty of Electronics, Telecommunications and Information Technology, University Politehnica of Bucharest, 2015.

List of publications

Journal papers

2021

• Alexandru Lucian Georgescu, Alessandro Pappalardo, Horia Cucu, Michaela Blott, “Performance vs. hardware requirements in state-of-the-art automatic speech recognition,” EURASIP Journal on Audio, Speech, and Music Processing 2021, no. 1 (2021): 28.

2018

• Alexandru Lucian Georgescu, Alexandru Caranica, Horia Cucu, Corneliu Burileanu, “RoDigits – a Romanian connected-digits speech corpus for automatic speech and speaker recognition,” in University “Politehnica” of Bucharest Scientific Bulletin, Series C, vol. 80, issue 3, pp. 45-62, Bucharest, 2018, ISSN: 2286-3540.

Conference papers

2021

• Alexandru-Lucian Georgescu, Horia Cucu, Corneliu Burileanu, “Improvements of SpeeD’s Romanian ASR system during ReTeRom project,” in the Proceedings of the 11th Conference on Speech Technology and Human-Computer Dialogue (SpeD), Bucharest, Romania, 2021.
• Alexandru-Lucian Georgescu, Cristian Manolache, Dan Oneaţă, Horia Cucu, and Corneliu Burileanu. “Data-Filtering Methods for Self-Training of Automatic Speech Recognition Systems.” In 2021 IEEE Spoken Language Technology Workshop (SLT), pp. 1-7. IEEE, 2021.

2020

• Cristian Manolache, Alexandru-Lucian Georgescu, Horia Cucu, Verginica Barbu Mititelu, and Corneliu Burileanu. “Improved text normalization and language models for SpeeD’s Automatic Speech Recognition System.” In the Proceedings of the 13th International Conference “Linguistic Resources and Tools for Processing the Romanian Language”, ConsILR 2020, Bucharest, Romania.
• Alexandru-Lucian Georgescu, Horia Cucu, Andi Buzo, Corneliu Burileanu, “RSC: A Romanian Read Speech Corpus for Automatic Speech Recognition“, in the Proceedings of the 12th International Conference on Language Resources and Evaluation (LREC), Marseille, France, 2020.
• Cristian Manolache, Alexandru-Lucian Georgescu, Alexandru Caranica, Horia Cucu, “Automatic Annotation of Speech Corpora using Approximate Transcripts,”  in the Proceedings of the 43rd International Conference on Telecommunications and Signal Processing (TSP), 2020, Milan, Italy.
• Dan Oneaţă, Alexandru-Lucian Georgescu, Horia Cucu, Dragoș Burileanu, Corneliu Burileanu, “Revisiting SincNet: An Evaluation of Feature and Network Hyperparameters for Speaker Recognition,” in the Proceedings of the 28th European Signal Processing Conference (EUSIPCO), Amsterdam, The Netherlands, 2020.

2019

• Florin Iordache, Alexandru-Lucian Georgescu, Dan Oneaţă, Horia Cucu “Romanian Automatic Diacritics Restoration Challenge“, in the Proceedings of the 14th International Conference on Linguistics Resources and Tools for Natural Language Processing, Cluj-Napoca, Romania, 2019.
• Alexandru-Lucian Georgescu, Horia Cucu, Corneliu Burileanu, “Kaldi-based DNN architectures for speech recognition in Romanian,” in the Proceedings of the 10th Conference on Speech Technology and Human-Computer Dialogue (SpeD), Timișoara, Romania, 2019.
• Alexandru-Lucian Georgescu, Horia Cucu, Corneliu Burileanu, “Progress on automatic annotation of speech corpora using complementary ASR systems,”  in the Proceedings of the 42nd International Conference on Telecommunications and Signal Processing (TSP), 2019, Budapest, Hungary.

2018

• Alexandru-Lucian Georgescu, Horia Cucu, Corneliu Burileanu, “Comparison of i-vector and GMM-UBM speaker recognition on a Romanian large speech corpus“, in the Proceedings of the 13th Edition of the International Conference on Linguistic Resources and Tools for Processing Romanian Language (ConsILR-2018), Iași, Romania.
• Alexandru-Lucian Georgescu, Horia Cucu, “Automatic annotation of speech corpora using complementary GMM and DNN acoustic models,”  in the Proceedings of the 41st International Conference on Telecommunications and Signal Processing (TSP), 2018, Athens, Greece.
• Alexandru-Lucian Georgescu, Horia Cucu, “GMM-UBM modeling for speaker recognition on a Romanian large speech corpora,” in the Proceedings of the 12th Romanian International Conference on Communications (COMM), 2018, Bucharest, Romania.
• Alexandru Caranica, Lucian Georgescu, Alexandru Vulpe, Horia Cucu, “Multilingual Low-Resourced Prototype System for Voice-Controlled Intelligent Building Applications,” in Rocha Á., Adeli H., Reis L., Costanzo S. (eds) Trends and Advances in Information Systems and Technologies. WorldCIST’18 2018. Advances in Intelligent Systems and Computing, vol 747. Springer, Cham.

2017

• Alexandru-Lucian Georgescu, Horia Cucu, Corneliu Burileanu, “SpeeD’s DNN Approach to Romanian Speech Recognition,” in the Proceedings of the 9th Conference on Speech Technology and Human-Computer Dialogue (SpeD), Bucharest, Romania, 2017, 8p, ISBN 978-1-5090-6496-0.

Dissertation Thesis

Alexandru Lucian Georgescu, “Automatic Speech Recognition System in Romanian Language based on Deep Neural Networks“, Dissertation Thesis, University Politehnica of Bucharest, 2018 (scientific coordinators: Assoc. Prof. Horia Cucu), graded 10/10 [demo].

Bachelor Thesis

Alexandru Lucian Georgescu, “Software applications in spoken language technology”, Bachelor Thesis, University Politehnica of Bucharest, 2016 (scientific coordinators: Lect. Horia Cucu, Prof. Corneliu Burileanu), graded 9.70/10 [demo].