Research Sciences Intern - Audio and Acoustics

Required/Minimum Qualifications:   

Must be currently enrolled in a PhD program in audio signal processing, acoustics, or a related machine learning field, including but not limited to, speech enhancement, speaker recognition, self-supervised learning, and/or semi-supervised learning. 

Other Requirements:  

Deep experience in at least one of the following areas: 

• Classical signal processing theory (adaptive filters, detection and estimation theory) 

• ML and AI algorithms, approaches to audio signal processing 

• Microphone signal processing 

• Multi-modal (e.g., audio-visual) signal processing 

• Numerical linear algebra 

• Demonstrable experience developing, characterizing, and implementing signal processing algorithms. 

Preferred/Additional Qualifications: 

• Ability to clearly communicate what work you have done, why it was important, and how it was different from existing projects. 

• Ability to work in ambiguous uncharted areas and having the experience, creativity, and technical depth to identify technical gaps, acquire missing information, align requirements, and pick the right direction. 

• Ability to create, train, and optimize neural network architectures for audio (or audio-visual) applications such as speech enhancement, speaker recognition, echo cancellation, source localization, audio-visual speaker diarization and active speaker detection. 

• Proven understanding of deep learning tools such as PyTorch. 
• Fluency in Python and/or Matlab.