Towards a neurally driven vocal prosthesis in an animal model for human speech

ARNEODO, EZEQUIEL M.

Querétaro

Seminario; Seminario Institucional; 2020

Instituto de Neurobiología, UNAM

Neural prostheses hold the promise of restoring lost function for individuals with motor,speech, and language deficits, as well as to advance our understanding of how the brainenables complex behavior. While limb-based motor prosthetic systems are actively studiedand have yielded increasingly high performance, speech and language prosthesesdevelopment has been much more limited. We are developing a neurally driven avian vocalprosthesis, as a model that provides a much-needed complement to existing human subjectswork and has the potential to accelerate the development of speech prostheses.In this talk I will present advances towards a brain machine interface (BMI) for birdsong, apowerful animal model for vocal learning that shares many features with human speech. Weexploit the fact that the biomechanics of birdsong production are well understood; thisenables us to employ a biophysical model of the vocal organ that captures most of thecomplexity of the song and reduces it to a lower dimensional parameter space. Thisdimensionality reduction allows us to achieve realistic synthetic birdsong driven by neuralactivity with the aid of simple neural networks, that are plausible to implement in real time.Finally, I will describe the advances in the fabrication of devices that aim to solve anothercritical issue of bioprosthetic devices: the need for long-term stable recordings and high yieldin neural interfacing.