Recent research suggests that body cameras alone do not improve policing. However, when coupled with analytics, professional police behavior increases which leads to more positive outcomes. When there is mutual respect, civilians are more willing to comply with the police.
Unprofessional language is defined as the use of profanity, insults, and threats. The objective of this blog is to provide strong evidence in favor of automated BWC analytics.
Data Collection and Analysis
The Alameda PD data for this study was collected via an Application program interface from the department’s evidence management system. Videos delivered through the API were fed through an audio extraction pipeline. This ensures only the audio information of the video is retained in temporary memory for analysis.
Video files containing only noise or silence were discarded during the audio analysis process. The audio analysis process involved feeding audio through machine-learned speech recognition models. Doing so obtained words and timestamps of valid speech in the audio file.
Results of training and automated BWC analytics:
36% decrease in the use of force
17% increase in officer explanations
12% decrease in civilian non-compliance
30% decrease in unprofessional officer language
99% of interactions were at a standard or high professionalism level
In parallel, this study observed a 30% decrease in officer use of unprofessional language between H1 2021 and H1 2022, despite a 27% increase in civilian usage of unprofessional language comparing the same time periods.
Thank You For Your Time!
Our extensive research confirms that the reduction of force occurrences improves officer professionalism. As a result, it improves civilian compliance, further proving that unprofessional language while attempting to de-escalate a situation does not reduce civilian non-compliance.
For those interested in the full case study, you can find it here.