Curriculum Vitae

Complete CV at Lattes platform (only in Portuguese)

Work Contact

Signals, Multimedia, and Telecommunications Lab.

Av. Horácio Macedo, 2030, room H-221 – Cidade Universitária

21941-598 Rio de Janeiro, RJ - Brazil

Tel/FAX.: +55 (21) 3938-8207

E-mail: igormq [at] poli [dot] ufrj [dot] br

Personal data

Brazilian, male, single, 25 years old.

Born in September 24, 1991 in Teresópolis, RJ - Brazil.

Name in bibliographic citations: QUINTANILHA, I. M.;QUINTANILHA, IGOR M.

Short biography

Igor Macedo Quintanilha received the B.Sc. degree in Electronics and Computing Engineering from Federal University of Rio de Janeiro (UFRJ), Brazil, in 2016. In 2012, he was a scientific initiation student at the Grupo de Teleinformática e Automação (GTA), focusing on vehicular networks. Since 2013, he has been working at the Laboratory of Signals, Multimedia, and Telecommunications (SMT). In 2017, he received his M.Sc. degree in Electrical Engineering from Federal University of Rio de Janeiro (UFRJ). He is currently working toward the D.Sc. degree with the Electrical Engineering Program (PEE) at Federal University of Rio de Janeiro, RJ, Brazil. He has interests in deep learning, speech recognition, computer vision, and signal processing.

Education

2017 - present

D.Sc. in Electrical Engineering

Federal University of Rio de Janeiro, UFRJ, Brazil

Title: TBD

Research area: Deep learning with applications in end-to-end automatic speech recognition for low-resource languagess

Advisors: Luiz Wagner Pereira Biscainho and Sergio Lima Netto

Keywords: lvcsr; connectionist temporal classification; deep learning

2015 - 2017

M.Sc. in Electrical Engineering

Federal University of Rio de Janeiro, UFRJ, Brazil

Title: All-neural speech recognition applied to Brazilian Portuguese

Research area: Deep learning with applications in end-to-end automatic speech recognition

Advisors: Luiz Wagner Pereira Biscainho and Sergio Lima Netto

Keywords: lvcsr; connectionist temporal classification; deep learning

2010 - 2015

B.Sc. in Electronic and Computing Engineering

Federal University of Rio de Janeiro, UFRJ, Rio de Janeiro, Brazil

Title: Algorithms for nonnegative matrix factorization with applications in percussive instruments transcription [text in Portuguese]

Advisor: Luiz Wagner Pereira Biscainho

Keywords: nmf; sound source separation

Complementary education

2016 - 2016

Machine Learning - online course offered through Coursera (Duration: 55h)

Stanford University, Stanford, USA.

2012

College extension in Introduction to Android Programming (Duration: 48h)

Federal University of Rio de Janeiro, UFRJ, Rio de Janeiro, Brazil.

2008

Student to Business (S2B) (Duration: 104h)

Microsoft Innovation Center, Microsoft, Brazil

Professional/Research Background

2020 - present

Founder & CTO at Octos Tecnologia LTDA.

Homepage: www.octos.ai

Activity: Reponsible for the tech area at Octos and the development of its core business.

2013 - 2015

Scientific Initiation project at Signals, Multimedia, and Telecommunications Lab (SMT)

Federal University of Rio de Janeiro, UFRJ, Rio de Janeiro, Brazil.

Research Area: Sound source separation

2013

Developer at Letsevo, Designoteca

Homepage: www.designoteca.com

Activities: I was responsible for changing the business logic of this startup. Improvements in user interface with Ruby on Rails and Javascript.

2012 – 2013

Scientific initiation project at Teleinformatics and Automation Group (GTA)

Federal University of Rio de Janeiro, UFRJ, Rio de Janeiro, Brazil.

Research Area: Vehicular network

Activities: I was in charge of developing a prototype of vehicle localization using wireless network (802.11n).

Research Projects

2017 - 2019

Research intern at Microsoft ATL Brazil team

Title: Deep learning applied to multimedia systems

Activities: Study, apply, and contribute with deep learning algorithms to multimedia systems, focusing on temporal tasks

2017 - 2018

Researcher in collaboration project

Federal University of Rio de Janeiro (SMT) / Siemens (Oslo, Norway)

Title: O&G Condition Monitoring: DP Vessel, Closed Ring Power System fault detection

Activity: Development of fault and anomaly detection algorithms in a closed-ring energy system for DP vessels.

2014

Researcher in collaboration project

Federal University of Rio de Janeiro (SMT) / Hewlett-Packard (HP Labs, Palo Alto)

Title: Multiple-Device Acoustic Localization (MeDAL)

Activities: Research and development towards new solutions for acoustic localization of sound sources and sensors.

Language

  • Portuguese: native
  • English: advanced

Research interests

  • Machine learning
  • Deep learning
  • Signal processing
  • Speech recognition

Publications

  1. Papers published in conferences
  • QUINTANILHA, I. M.; BISCAINHO, L. W. P. ; NETTO, S. L. . A New Automatic Speech Recognizer for Brazilian Portuguese Based on Deep Neural Networks and Transfer Learning. In: Congreso Latinoamericano de Ingeniería de Audio de la AES, 2018, Montevidéu. Memorias del AES LAC 2018, 2018. p. 16-22.
  • PASSOS, W. L. ; QUINTANILHA, I. M. ; ARAUJO, G. M. . Real-Time Deep-Learning-Based System for Facial Recognition. In: XXXVI Simpósio Brasileiro de Telecomunicações e Processamento de Sinais, 2018, Campina Grande, PB. Anais do XXXVI Simpósio Brasileiro de Telecomunicações e Processamento de Sinais, 2018. p. 895-899.
  • QUINTANILHA, I. M.; BISCAINHO, L. W. P. ; NETTO, S. L. . Assessing the Performance of a Post-Processed End-to-End Speech Recognizer for Brazilian Portuguese Using Deep Neural Networks. In: Congresso de Engenharia de Áudio da AES-Brasil, 2017, Florianópolis. Anais do 15o. Congresso de Engenharia de Áudio da AES-Brasil, 2017. p. 55-62.
  • QUINTANILHA, I. M.; BISCAINHO, L. W. P. ; NETTO, S. L. . Towards an End-to-End Speech Recognizer for Portuguese Using Deep Neural Networks. In: Simpósio Brasileiro de Telecomunicações e Processamento de Sinais, 2017, São Pedro. Anais do XXXV Simpósio Brasileiro de Telecomunicações e Processamento de Sinais, 2017. p. 408-412.
  • RIBEIRO, J. G. ; QUINTANILHA, I. M. ; CAMPISTA, MIGUEL ELIAS M. ; COSTA, LUIS HENRIQUE M. K. . Sistema para Monitoramento Descentralizado de Trânsito baseado em Redes Veiculares Infraestruturadas. In: XXXI Simpósio Brasileiro de Redes de Computadores e Sistemas Distribuídos, 2013, Brasília. Trilha Principal e Salão de Ferramentas, 2013. p. 863-876. [paper in Portuguese]
  • RIBEIRO JUNIOR, JOSE GERALDO ; QUINTANILHA, IGOR M. ; CAMPISTA, MIGUEL ELIAS M. ; COSTA, LUIS HENRIQUE M. K. . Evaluation of an opportunistic collaborative traffic monitoring system. In: 2012 IFIP Wireless Days (WD), 2012, Dublin. 2012 IFIP Wireless Days. p. 1-6.
  1. Technical productions
  • QUINTANILHA, I. M.; CINELLI, L. P. . Introduction to Deep Learning and its Applications. 2018. (Short duration course, in Portuguese).
  • AMORIM, B. D. C. G. ; QUINTANILHA, I. M. . Course of Web Development in Ruby on Rails. 2013.
  • QUINTANILHA, I. M. ; ESTEVÃO FILHO, R. M. Introduction to Deep Learning with Applications. 2016.

Organization of events, conferences, and expos

  • BISCAINHO, L. W. P. ; MANNIS, J. A. ; ANSELMO, J. ; BRITO, J. ; QUINTANILHA, I. M. ; APOLINARIO, I. F. ; COSTA, M. V. M. . Regional Meeting of AES-Brazil. 2016.