Read, code, debug, repeat.

Hi, I’m Lester Violeta, a Ph.D. student at Nagoya University, Japan at Toda Laboratory under the supervision of Professor Tomoki Toda through a full scholarship from the Monbukagakusho Japanese Government. My recent research mainly focuses on speech synthesis, particularly with electrolaryngeal speech data and singing voice data, where my work has been published in top speech and audio conferences such as Interspeech, ICASSP, ASRU, SLT, and TASLP. I was co-organizer of the recent Singing Voice Conversion Challenge in 2023. I am also part of the review committee for several academic conferences such as SLT, ICASSP, and Interspeech. Aside from research, I am also currently a founding AI engineer at VoiceSwap.AI. Previously, I also worked on research internships at Sony Computer Science Laboratories (Tokyo, Japan), NTT Media Intelligence Laboratories (Kanagawa, Japan), and Hitachi Ltd (Tokyo, Japan). Thus, I have extensive experience in both the academic research and engineering sides of AI.

Prior to doing research, I received my B.S. in Electronics Engineering degree from Ateneo de Manila University, Philippines. Moreover, I was also a research exchange student at Institut Catholique d’Arts et Metiers - Site de Paris-Senart in France where I worked on optimizing renewable energy systems for a semester during my undergraduate degree. Some of my hobbies include language learning, where I learned French during my exchange program and am currently studying Japanese. Having this hobby led me to gain an interest in research in speech and languages, so here I am!

For more details, you can download my CV here or check out my papers in my Google Scholar here.

Updates

  • Oct 2024: I made a demo system with a frontend for my main thesis and a side project, check them out! Electrolaryngeal Speech Enhancer and Singing Voice Converter (in Japanese).
  • May 2024: First-author journal on pretraining and adaptation techniques for electrolaryngeal speech recognition. Accepted at IEEE TASLP/ACM.
  • May 2024: Co-authored paper on speaker verification for pathological speech. Accepted at INTERSPEECH 2024 (Kos, Greece).
  • Feb 2024: Started as an AI Engineer / Researcher in Voice-Swap.AI
  • Dec 2023: First-author paper on electrolaryngeal speech intelligibility enhancement. Accepted at ICASSP 2024 (Seoul, South Korea).
  • Oct 2023: Started as a research assistant at Sony CSL in Japan.
  • Sept 2023: Three co-authored papers accepted at ASRU 2023 (Taipei, Taiwan). Check them out here: The Singing Voice Conversion Challenge 2023, Detailed study on training singing voice conversion systems, and Healthy to pathological voice conversion using style tokens.
  • Aug 2023: First-author paper on speech recognition for deaf and hard-of-hearing speakers. Accepted at APSIPA 2023 (Taipei, Taiwan).
  • Apr 2023: Started my Ph.D. degree at Nagoya University.
  • Feb 2023: First-author paper on using data augmentation for electrolaryngeal speech recognition. Accepted at ICASSP 2023 (Rhodes, Greece).
  • Jan 2023: I will be co-organizing the Singing Voice Conversion Challenge! More details can be found here. Please join!
  • Sep 2022: Co-authored paper on electrolaryngeal speech intelligibility enhancement. Accepted at SLT 2022 (Doha, Qatar).
  • May 2022: First-author paper on analyzing self-supervised speech models for pathological speech recognition. Accepted at INTERSPEECH 2022 (Incheon, Korea).
  • Mar 2022: Accepted as a research intern at NTT Japan.
  • Jan 2022: Co-authored paper on pathological voice conversion. accepted at ICASSP 2022 (Singapore, Singapore).
  • Jan 2022: Accepted as a research intern at Hitachi Ltd.
  • Apr 2021: Started my master’s degree at Nagoya University.