I'm developing a pipeline to sync underwater passive acoustic audio with whale sightings around the hydrophones, to then classify that audio using Google's open sourced whale models. The goal is to enrich data from happywhale.com with whale voices, so scientists can further explore communication of these species. I'm trying to keep things as system-agnostic as possible, but am building the first implementation to run on GCP.
https://github.com/pmhalvor/whale-speech