HLT@INESC-ID

Sérgio Ferreira Paulo graduated in Electrical and Computer Engineering in 2001 from Instituto Superior Técnico (IST), Lisbon. Then, he joined the Spoken Language Systems Labs of INESC-ID for a PhD on "Automatic Methods for Building Speech Synthesis Corpora", during which he was involved in several national and international research projects. In 2009, he received a Doutor degree in Electrical and Computer Engineering from IST of the Technical University of Lisbon. His research interests span from optimal corpus design to automatic phonetic segmentation, with a strong emphasis on pronunciation modeling, applied to speech synthesis.

Publications

International Conferences

2022

• Anna Havras, Carlos Mendes, G. Hristovsky, Sérgio Paulo, Helena Moniz, Speech Technologies and Russian Pronunciation Variation in the context of an international Artificial Intelligence (AI) company, In YFRSW 2022, ISCA, Incheon, South Korea, September 2022

2008

• Sérgio Paulo, Luís C. Oliveira, Carlos Miguel Duarte Mendes, Luís António Dias Mestre Figueira, Renato Márcio de França Cassaca, Maria do Céu Guerreiro Viana Ribeiro, Helena Moniz, DIXI - A Generic Text-to-Speech System for European Portuguese, In PROPOR 2008 - 8th International Conference on Computational Processing of the Portuguese Language, Springer-Verlag, vol. 5190, series LNAI, pages 91-100, Aveiro, Portugal, September 2008

• Luís C. Oliveira, Sérgio Paulo, Luís António Dias Mestre Figueira, Carlos Miguel Duarte Mendes, The INESC-ID Blizzard Entry: Unsupervised Voice Building and Synthesis, In Blizzard Challenge 2008, Brisbane, Australia, September 2008

• Luís C. Oliveira, Sérgio Paulo, Luís António Dias Mestre Figueira, Carlos Miguel Duarte Mendes, Ana Nunes, Joaquim Godinho, Methodologies for Designing and Recording Speech Databases for Corpus Based Synthesis, In LREC 2008, The sixth international conference on Language Resources and Evaluation, May 2008

2007

• Christian Weiss, Sérgio Paulo, Luís António Dias Mestre Figueira, Luís C. Oliveira, Blizzard Entry: Integrated Voice Building and Synthesis for Unit-Selection TTS, In Blizzard 2007, ISCA, Bonn, August 2007

• Christian Weiss, Luís C. Oliveira, Sérgio Paulo, Carlos Miguel Duarte Mendes, Luís António Dias Mestre Figueira, Marco Vala, Pedro Sequeira, Ana Paiva, Thurid Vogt, Elisabeth Andre, ECIRCUS: Building Voices for Autonomous Speaking Agents, In 6th Speech Synthesis Workshop, ISCA, Bonn, August 2007

• Sérgio Paulo, Luís C. Oliveira, MuLAS: A Framework For Automatically Building Multi-Tier Corpora, In Interspeech, ISCA, Antwerp, August 2007

2005

• Sérgio Paulo, Luís C. Oliveira, Generation of Word Alternative Pronunciations Using Weighted Finite State Transducers, In Interspeech'2005, ISCA, pages 1157-1160, Lisbon, Portugal, September 2005

• Sérgio Paulo, Luís C. Oliveira, Reducing the Corpus-based TTS Signal Degradation Due to Speaker's Word Pronunciations, In Interspeech'2005, ISCA, pages 1089-1092, Lisbon, Portugal, September 2005

2004

• Sérgio Paulo, Luís C. Oliveira, Automatic Phonetic Alignment and Its Confidence Measures, In 4th International Conference EsTAL 2004, Springer, series Lecture Notes in Artificial Intelligence, pages 36-44, Alicante, Spain, October 2004

2003

• Sérgio Paulo, Luís C. Oliveira, DTW-based Phonetic Alignment Using Multiple Acoustic Features, In EUROSPEECH'2003 - 8th European Conference on Speech Communication and Technology, Genéve, Switzerland, September 2003

• Isabel Trancoso, Maria do Céu Guerreiro Viana Ribeiro, Manuela Barros, Diamantino António Caseiro, Sérgio Paulo, From Portuguese to Mirandese: fast porting of a letter-to-sound module using FSTs, In PROPOR'2003 - 6th International Workshop on Computational Processing of the Portuguese Language, Springer, series Lecture Notes in Artificial Intelligence, June 2003

• Sérgio Paulo, Luís C. Oliveira, Improving the Accuracy of the Speech Synthesis Based Phonetic Alignment Using Multiple Acoustic Features, In Computational Processing of the Portuguese Language - Proc. of the 6th Intl. Workshop, PROPOR 2003, Springer-Verlag, Heidelberg, series Lecture Notes in Artificial Inteligence, pages 31-39, Faro, Portugal, June 2003

2002

• Sérgio Paulo, Luís C. Oliveira, Multilevel Annotation Of Speech Signals Using Weighted Finite State Transducers, In 2002 IEEE Workshop on Speech Synthesis, Santa Monica, California, September 2002

Technical Reports

2008

• David Martins de Matos, Ricardo Ribeiro, Sérgio Paulo, Fernando Batista, Luísa Coheur, Joana Paulo Pardal, Natural Language Engineering on a Computational Grid (NLE-GRID) T2 - Encapsulation of Reusable Components, Tech. Rep. 31 / 2008 INESC-ID Lisboa, January 2008

• Tiago Luís, David Martins de Matos, Sérgio Paulo, Ricardo Ribeiro, Natural Language Engineering on a Computational Grid (NLE-GRID) T5 - Performance Experiments, Tech. Rep. 35 / 2008 INESC-ID Lisboa, January 2008

Doctoral Theses

2009

• Sérgio Paulo, Automatic Methods for Building Speech Synthesis Corpora, PhD thesis, Instituto Superior Técnico, Lisbon, June 2009

Research Interests

• Speech synthesis
• Automatic corpora building techniques
• Speech processing