DEPARTMENT INVESTS IN
Alameda Police Chief Nishant Joshi had only been on the job for six months and made acquiring the at-risk language detection system one of his top priorities. He explained how it works.
“This technology points directly to sentiment within the conversation. It identifies good interactions and it identifies at-risk interactions. We can also [add], we want to know every time our officer says a phrase something like, is there anything else I can do to support you,” Chief Joshi said.
Chief Joshi says the goal is to use the audio analyzing technology as a training tool to highlight excellent police service and point out any potential behavioral issues that should not be modeled by his officers when interacting with the public.
With 90 officers equipped with cameras on the frontlines, the quantity of footage is too large for manual review
Large quantity of footage without automated review poses risk to miss potential behavioral issues as well as examples of excellent service
Challenging or impossible to find specific examples of phrases and language used such as “how else can I support you” across large data set
Alameda is uploading and processing 1,000 hours of BWC audio through Truleo for:
Auto - classification of events contained in footage.
Customization of reports for Alameda's specific needs (i.e., de-escalation tactics and respectfulness)
Human level accuracy at scale.
Cost & time efficiency savings.
Streamlined & substantiated risk management processes.
Identification of training tactics and best practices in policing