sounds as sources of intelligence
OpenEars is the open-source version of Serval.
Want to see how the city of Amsterdam sounds like?
Look here right now!
The Serval sound sensor is being developed to monitor nature reserves. Poachers follow logical routes to get to their prey. There are a few critical tracks, a few places where they can cross the river. Sensors placed at such strategic locations can pick up the sound of passing cars or motorboats at those strategic places. Once that happens park managers are notified and can take appropriate action.
The Serval sound sensor can be seen as an intelligent microphone. It is attached to a small computer that hosts a trained neural network able of classifying captured sounds. It does so in real time and upon the sensor, and sends out the classified sounds. This way park managers can remotely hear what sounds are being heard deep in the forest.
Neural network
The neural network is a complex model created by the Sensing Clues team. It indicates, for example, that it hears a car with eighty percent certainty. But it immediately adds that it can also be an motorbike. That probability is twenty percent. It is up to the receiver to make a good estimate of the situation. He has to decide: if we observe human activity of this type at that time and place, we have to go for it.
Open Source
We open-sourced Serval under the name OpenEars. If you are interested in it or want to be involved in its development, don’t hesitate to contact us.
serval in 90 seconds
90 seconds explanation and demonstration of the sensor
