What Is a Cognitive Engine?
A cognitive engine is an AI application that ingests unstructured data and analyzes it, then produces a result in a structured form.
There are cognitive engines for a wide range of different functions. For example, in the realm of text analysis, there are engines for language identification, translation, content classification, summarization, sentiment analysis and more. For analyzing video, engines exist for object detection, license plate recognition and optical character recognition. For speech, engines can be used to convert speech in audio or video files into text transcripts, identify speakers based on recordings of their voices and break audio files down into segments that separate out different speakers.
One of the leading providers of AI engines is a company called Veritone, which has an operating system for AI, aiWARE. The software “orchestrates a diverse ecosystem of machine learning models to transform audio, video, text, and other data sources into actionable intelligence,” according to the company’s website.
Veritone notes the platform “uses machine learning across the AI engine ecosystem to orchestrate and employ the best engines for the job, always producing optimal results.”
In the public safety context, such a solution has a variety of useful applications. One is to enable law enforcement to more quickly release to the public footage from body cameras, dash cameras or surveillance cameras, while redacting sensitive information such as the identity of innocent individuals or information that would compromise an ongoing investigation.
Such a tool allows public safety agencies to more quickly respond to public calls for the release of footage of interactions with citizens. Instead of having users engage in a time-consuming manual redaction process, the software can detect human faces and lets users define other sensitive imagery, according to Veritone. The tool then automatically redacts it from audio-, video- and image-based evidence.
LEARN MORE: Follow these best practices when deploying a next-generation work center.
The Benefits of Cognitive Engines for Public Safety
There are several ways public safety agencies benefit from employing cognitive engine technology in RTCCs. One is obviously increased transparency. The more quickly evidence can be analyzed and redacted appropriately, the more quickly in can be released to the public.
Delays in releasing such evidence can sometimes lead to the perception that an agency is hiding something. Expediting the review and release of evidence can increase public trust and strengthen the relationship between a law enforcement agency and the community it serves.
“As law enforcement agencies are under increasing scrutiny, we welcome more transparency. We want to be able to share relevant videos proactively with the public, particularly those from body cams,” says David Jantas, the former chief of police at New Jersey’s Pemberton Township Police Department, on Veritone’s site. “To make this happen, we needed sophisticated redaction software. Veritone quickly emerged as the right partner –– its Redact solution checked all of our boxes for security, including CJIS compliance and the stability of Microsoft Azure’s cloud. It’s also saving us taxpayer dollars, resources and time.”
Cognitive engines are also force multipliers. They help cut down on the resources and time required of staff to analyze evidence. Instead of focusing on time-intensive but low-value tasks such as redaction and transcription, personnel can focus on higher-value work.
At a time when public safety agencies are facing greater demands for transparency from the public, technology tools can help them meet those demands while serving their communities.
This article is part of StateTech’s CITizen blog series. Please join the discussion on Twitter by using the #StateLocalIT hashtag.
