How to Apply
- Resume (1 page max): Highlight relevant experience and skills in computer science, engineering, and mathematics.
- Optional GitHub Submission:
- Task: Build a basic speech-to-text system from scratch (without external libraries for ASR such as Speech Recognition, google-speech, etc.). Your system should:
- Preprocess audio and extract features (e.g., MFCCs, or using a method like Wav2Vec).
- Implement a simple recognition method (e.g., HMM, or any end-to-end approach).
- Evaluate performance on a small dataset.
- Ideal Submission: A clean, short, and well-documented GitHub repository with a brief README.md explaining your approach. Include a link to your repository with your application.
- Include link to submission within the body of the resume
Shortlisted candidates will receive an email with details about the next steps in the application process.
Job Summary
Department: LSA Linguistics
Institution: University of Michigan, Ann Arbor
Job Description: We are seeking motivated University of Michigan student interns to support research projects focused on language models and automatic speech recognition (ASR). This role offers a unique opportunity to gain hands-on experience with cutting-edge machine learning and natural language processing (NLP) technologies. Student interns will engage in various coding tasks and contribute to the development and refinement of state-of-the-art models.
What You'll Get:- Hands-on experience with the latest NLP and ASR technologies
- Access to powerful computers, including GPU hours in the Great Lakes High-Performance Computing (HPC) cluster.
- Access to popular datasets like TIMIT and Librispeech.
- Access to the labs data storage space.
- Strong recommendation letter for the application of any positions
- Salary: $20-$25 / hour, depending on skill level
Required Qualifications*
- We welcome University of Michigan undergraduate or graduate students from any major (like computer science, linguistics, engineering, mathematics, cognitive science, etc.).
- You should know how to code in Python.
- You should be familiar with machine learning techniques (like Transformers, LSTM).
- You should be able to explain your ideas clearly and work well with others (see Amazon Leadership Principles for a specific guide).
Desired Qualifications*
- Coursework or projects related to ASR, NLP, or general machine learning, including online courses like Coursera.
- Experience using deep learning tools like PyTorch or TensorFlow.
- Experience building websites.
Work Schedule
Hours: estimated 10 hours per week, with flexibility depending on the project and your schedule.
Modes of Work
Positions that are eligible for hybrid or mobile/remote work mode are at the discretion of the hiring department. Work agreements are reviewed annually at a minimum and are subject to change at any time, and for any reason, throughout the course of employment. Learn more about the work modes.
Application Deadline
Applications are reviewed on a rolling basis until filled.
U-M EEO/AA Statement
The University of Michigan is an equal opportunity/affirmative action employer.