Details
LINK TO ZOOM MEETING:
https://us06web.zoom.us/j/88384219320?pwd=dlR1WHpBM1JJT2ZXdmE5dSthRWp1UT09
PROGRAM DESCRIPTION: CETI (Cetacean Translation Initiative), a multidisciplinary research team combining linguistics, marine biology, and machine learning, used AI to analyze the clicking sounds (or codas) from hydrophone-tag recordings of 9,000+ sperm whales off the Caribbean island of Dominica. The recordings captured thousands of codas exchanged during social interactions—sometimes head-to-head “conversations.”
The analysis indicates that the sounds weren't simple animal sounds or random clicks, but were linguistically-based communications consisting of a phonetic alphabet and even vowel-like elements, making them one of the closest parallels to human language ever found in another species.
AI models like GANs and deep acoustic classifiers were used to map the codas into a symbolic system, much like decoding an ancient language. Linguist Gašper Beguš compares the process to reconstructing lost human tongues separated not by time but by environment—“oceans apart.”
Tune in to Wednesday's program and discussion that describes sperm whale behavior and how they communicate.
MEET A LEADER IN THE FIELD: Gašper Beguš, the head of the CETI (Cetacean Translation Initiative) research team, investigates how intelligence—human, animal, and artificial—unfolds through language. He is an Associate Professor of Linguistics at UC Berkeley, where he focuses on interpretable AI and combines linguistics, cognitive science, machine learning, neuroscience, philology, and marine biology.
FURTHER LOGIN INFO:
LINK: https://us06web.zoom.us/j/88384219320?pwd=dlR1WHpBM1JJT2ZXdmE5dSthRWp1UT09https://us06web.zoom.us/j/88384219320?pwd=dlR1WHpBM1JJT2ZXdmE5dSthRWp1UT09https://us06web.zoom.us/j/88384219320?pwd=dlR1WHpBM1JJT2ZXdmE5dSthRWp1UT09
DIAL-IN PHONE NUMBER: 305-224-1968
MEETING ID: 883 8421 9320
PASSCODE: 765292