Applying machine learning for detecting vocalizations
Michael Yartsev, Professor
Bioengineering
Closed. This professor is continuing with Fall 2023 apprentices on this project; no new apprentices needed for Spring 2024.
This project aims at describing the vocal behavior of the Egyptian fruit-bat. This social species of bats is known to emit a diversity of vocalizations in the wild, but its repertoire and vocal behavior still remains to be carefully described. At the Yarstev Lab we are particularly interested in understanding the differences in vocal productions between sexes and across ages to highlight the ontogeny of vocal production and differences in usage. To explore the ontogeny of vocal production in particular, we are testing if auditory input is necessary for normal vocal development in young animals. If vocal development is affected by the lack of auditory feedback (hearing self-vocalizations) and auditory input (hearing others’ vocalizations), this would be a good clue that a learning process is engaged for the production of vocalizations. In other words we are investigating whether vocalizations are innate, or acquired through a learning process.
Role: Audio recordings from several groups of bats (families and experimental animals) were obtained in the lab and a fair portion of the data has been manually labeled. The student will apply its knowledge in machine learning technics (neural networks in particular) to code a matlab algorithm that is trained on the manually curated data and detects vocalizations in the other recordings. The student will remotely connect to computers in the lab to access data and elaborate the code. The students will learn how sound is represented and analyzed, and state of the arts technics that are used to record and extract bioacoustics features.
Qualifications: Successful research apprentices complete their assignments in a timely manner, maintain open communication with other members of the research group and with the research coordinator, ask questions when they need help or guidance, and actively ensure (through communicating with the research coordinator) that they are getting the experience they want from the URAP program. Good communication skills and knowledge in coding and machine learning are required, knowledge of sound physics is desirable but not essential.
Day-to-day supervisor for this project: Julie E Elie, Staff Researcher
Hours: to be negotiated
Off-Campus Research Site: Can be entirely remote
Related website: https://www.researchgate.net/profile/Julie-Elie/research
Related website: https://www.researchgate.net/profile/Julie-Elie/research