In a collaboration between UltraVoice, Ltd. and the Kording lab at UPenn we want to evaluate how much we can improve the voices of people who have lost their natural voice box (larynx) using deep learning.

The approaches would include semi-supervised learning, a new kind of speech data, and a problem that is similar to speech to speech. This is an advanced deep learning project.

Requirements: Experience with Pytorch or Tensorflow is essential, experience with speech data a major plus.

The successful applicant will develop experience in both industry (Dr. David Baraff) and academia (supervised by Dr. Konrad Kording), with an opportunity to collaborate on publications stemming from the the project.

The applicant may work remotely, but candidates must be US CITIZENS. Pay is commensurate with experience.

Applications will be reviewed on a rolling basis, with decisions made by May 1, 2020.

To apply, candidates should submit a brief Cover Letter and CV to Professor Konrad Kording (kording@upenn.edu).