Postdoctoral Researcher – Speech and NLP

Postdoc with speech and NLP expertise sought to join a team using Generative AI, Machine Learning, and Learning Engineering to study and improve digital tools to improve reading instruction for elementary school English learners.  This collaborative project spans Penn’s Linguistic Data Consortium, the Department of Computer and Information Science, the Penn Center for Learning Analytics, Digital Promise, and Amira Learning. We will develop better methods for linguistic adaptation, personalization, and disengagement intervention and test them on real students. Our goal is to advance methods in these areas in ways that improve our ability to teach students to read. 

The postdoc opening is in collaboration between Mark Liberman, Lyle Ungar, Neville Ryant, and Ryan Baker.

Skills required or preferred:

• Required: experience in machine learning, natural language processing, learning analytics, or affective computing
• Required: solid Python and PyTorch programming skills 
• Required: ability to work closely with project partners across multiple disciplines in education and computer science
• Preferred experience with:
  • pre-training and fine-tuning speech representation learning models such as wav2vec 2.0, Hubert, unispeech-SAT, and WavLM
  • speaker embeddings, diarization, and continuous speech separation
  • affect detection or disengagement detection
  • digital learning platforms and learning engineering

Please upload CV and cover letter to https://docs.google.com/forms/d/e/1FAIpQLSf-W8F0iouH3_tuOhju2RJeF9sdxk5wh2iqIpSBGP18S1aVdw/viewform?usp=sf_link by December 11, 2024.

The University of Pennsylvania values diversity and seeks talented students, faculty and staff with diverse backgrounds, experiences, and perspectives. The University of Pennsylvania does not discriminate on the basis of race, color, sex, sexual orientation, gender identity, religion, creed, national or ethnic origin, citizenship status, age, disability, veteran status or any other legally protected class status in the administration of its admissions, financial aid, educational or athletic programs, or other University-administered programs or in its employment practices.