To deliver meaningful ocean metrics from real-time and archived acoustic data sets, SoundSpace Analytics develops modular algorithms and fully automated processing pipelines that work hand-in-hand to move acoustic data from sea to screen.
- Bio-acoustic call detection
Most marine species use sound to navigate their world. Often, acoustic activity is used to infer habitat usage, and in many cases, behavioral cues are revealed. SoundSpace algorithms detect and classify vocalizations from many marine species using Artificial Intelligence. Custom training libraries are used to tune the software for local soundscapes.
- Source localization and tracking
For data from hydrophone arrays, SoundSpace provides software that can localize whale vocalizations in near real-time. Subsequently, acoustic whale tracks are obtained through algorithmic trajectory computations to infer the direction of their movement and speed.
- Acoustic vessel detection
Understanding the contribution of vessel noise to the soundscape is critical to measure acoustic ocean health. SoundSpace AI-trained acoustic vessel detectors exploit acoustic data to identify the presence of vessel noise. The detectors can capture engine powered vessels of all sizes and don't rely on AIS.
The vessel detectors are developed in collaboration with Ocean Acoustics Ltd. and trained with a broad range of vessel signatures collected globally in different soundscapes.
- Sound level analysis
Any recorded soundscape contains elements from environmental, anthropogenic, and bionic sources, often mixed with unwanted system noise. The detailed composition of a soundscape and temporal trends tell a compelling story with respect to ocean health and thus inform management decisions. SoundSpace Analytics provides the software to process, analyze, and visually present customized sound level metrics.
- Data visualization
SoundSpace Analytics provides the infrastructure to produce static or interactive visualizations for any custom metrics. Cutting the edge for data visualization, I co-developed SeaStats, an Open Source web dashboard for instant and interactive availability of soundscape metrics.