Morgan Lee Sandler

Applied Research and Data Scientist

Computer Science & Engineering
428 S. Shaw Lane, Rm 3115
East Lansing, Michigan, USA

About me

I am a master’s student in the Integrated Pattern Recognition and Biometrics Laboratory (iPRoBe) located at Michigan State University. I am under the supervision of Dr. Arun Ross. My latest resume can be downloaded here. Feel free to contact me via email.

Research Interests

Speech and Signal Processing, Speaker Recognition, Biometrics, Machine Learning

Technical Skills

Python (5 years exp.), SQL (2 years exp.), TensorFlow (3 years exp.), PyTorch (3 years exp.), Scikit-Learn (3 years exp.), C++ (3 years exp.), Research, Scientific Writing

News

Poster presentation @ 2nd Data Science Student Conference, November 2022
Poster presentation @ MSU Graduate Engineering Research Symposium, August 2022
Poster presentation @ MID-SURE Undergraduate Research Symposium, July 2021

Highlighted Experiences (Full Resume)

Raytheon Technologies, Data Scientist Internship, Summer 2022

Established a 185,000-node graph based upon employee-management hierarchy in Python, pandas, and NetworkX to model the spatial relations of employee behaviors and attitudes for visualization of company-wide (post-merger) user adoption of Office 365 (O365) tools.

Pratt & Whitney, Software Engineering Internship, Summer 2021

Built a data entry system in ASP.NET, EF Core, and C# to allow secure access to critical SQL data used by 100+ senior managers in vital business decisions.

United Airlines, Machine Learning Capstone Project, Spring 2021

Built a Computer Vision software with Python, YOLOv3, and template matching to identify aircraft maintenance tasks for auditing safety risks during aircraft turns at Chicago O’Hare airport.

Quicken Loans, Software Engineering Internship, Summer 2020

Identified non-match errors in loan document processing software which affects 3% of documents. Built a solution that implements distance and cosine similarity algorithms in OpenEdge ABL to tolerate small input discrepancies.

Honors & Awards

Awarded the Christopher J. Jackson Endowed Scholar of Computer Science, 2021

Publications

Is style all you need? Dependencies between emotion and GST-based speaker recognition paper illustration

Sandler, M., Ross, A.
Is style all you need? Dependencies between emotion and GST-based speaker recognition

In Coming Soon,2022.

In this work, we study the hypothesis that speaker identity embeddings extracted from speech samples may be used for detection and classification of emotion. In particular, we show that emotions can be effectively identified by learning speaker identities by use of a 1-D Triplet Convolutional Neural Network (CNN) & Global Style Token (GST) scheme (e.g., DeepTalk Network) and reusing the trained speaker recognition model weights to generate features in the emotion classification domain.

Main URLarxiv.org