Restoring natural speech from whispered speech in real-time and on-the-fly: looking beyond the 80-year-old electrolaryn

When a speaker is unable to produce normal (voiced) speech due the absence or malfunction of the vocal folds, also referred to as vocal cords, whispered speech is the only option available for oral communication, which is problematic from multiple points of view. One possibility to overcome this difficulty is converting whispered speech to natural speech in real time, and on-the-fly, such that conversational interaction is not disrupted. The electrolarynx is such an assistive technology still being marketed today but it is more than 80-year-old, and produces highly unnatural, unpleasant and robotic speech. In this keynote/seminar, we will discuss the main theoretical concepts underlying the theme, and practical solutions to the problem, highlighting approaches that are based on the source-filter model of voice production, and that rely on parametric models of signal analysis and synthesis. Theory and results will be presented in an illustrative manner.


Biography:
Aníbal Ferreira (Eng. Lic. in 1988, Msc in 1992, PhD in 1998, Habilitation in 2023) is Associate Professor (with Habilitation) in the ECE Department of the Faculty of Engineering at the University of Porto, in Portugal, where he lectures in the areas of signal theory, physiological signal processing, multimedia and telecommunications.
His research career started in 1988 at Philips Research Labs, in Eindhoven (The Netherlands), on automatic VLSI silicon compilation of signal processing blocks.
In 1990/91, and in 1993, he was a consultant at AT&T Bell Laboratories Murray Hill, New Jersey, in the area of perceptual audio coding. This work led to relevant contributions to the specification of the MPEG-Audio AAC standard, and to proprietary solutions currently in use for satellite radio broadcasting (SiriusXm Satellite Radio).
Dr. Ferreira has participated in several European research projects and has coordinated 10 Portuguese research projects in the areas of real-time audio analysis, synthesis, compression, modification, transcription, and dysphonic voice analysis and reconstruction.
Dr. Ferreira has also been involved in several entrepreneurial initiatives addressing voice quality assessment, biofeedback in stuttering treatment, visual feedback of the singing voice, and multimedia communication.
His research interests include psychoacoustics, audio and voice/speech analysis, synthesis and coding, multirate filter banks, acoustic analysis of the spoken and singing voice, dysphonic voice reconstruction, and audio forensics.
As an appreciator of cultural (and organizational) diversity, Dr. Ferreira has visited several universities, namely in the context of sabbatical leaves, including Massachusetts Institute of Technology (MIT), Massachusetts, USA (2020), Georgia Institute of Technology (GaTech), Georgia, USA (2010), and Rutgers University, New Jersey, USA (2006).
His research interests include psychoacoustics, audio and voice/speech analysis, synthesis and coding, multirate filter banks, acoustic analysis of the spoken and singing voice, dysphonic voice reconstruction, and forensic audio.
He is a lover of Nature, appreciator of the various Arts, admirer of the human creative genius, and his favorite hobbies are sports (running, swimming, cycling), and drumming.