par Ryan Mccallum Il y a 3 années
141
Plus de détails
The Pedestrian Alert System would be an embedded application within mobile devices which would automatically alert users to the proximity of a car using ambient noise as an indicator. It would intake noise and learn how to distinguish between engine sounds and other ambient noise by running the sounds through a model trained to hear those kinds of sounds. From this, a distance will be estimated from the user and the application will attempt to determine if the user is at risk of bring hit by a car. In the case a user is determined to be in danger, provide haptic and aural feedback to alert them of an issue.
Model Training
Audio Storage
Cognitive Computing
Artificial Intelligence