Browse Content by Topic:
Automatically Search Audio Recordings for Key Spoken Words
Author: Tony Procops, CyberTech North America
Copyright: Copyright 9-1-1 Magazine, Feature Content
The Next Generation 9-1-1 (NG9-1-1) initiative seeks to significantly expand the capabilities of today’s emergency 9-1-1 centers in order to optimize call-taker performance and enhance the center’s overall incident response and resolution efforts. Primarily focused on a center’s ability to handle text, data, images and video, there is another significant component to also consider – the use of audio-based intelligence in the mission-critical emergency services environment.
With new advanced technologies on the market, 9-1-1 centers can instantly search recorded audio interactions (landline calls, cell phone calls and radio transmissions) for select spoken words to automatically and accurately:
- Sift through hordes of tip-line calls to identify that viable needle-in-a-haystack call – e.g. if the words “yellow scarf” are mentioned by a tipster on the telephone (landline or cell), and that aspect of a missing person case was never made public.
- Identify relationships between present and past incidents – e.g. if the same unique modus operandi of a killer was reported by first responders (over the radio) in two separate cases.
- Identify good and bad call-taker interactions to help with dispatcher assessment and training – e.g. a good call could be identified by picking up the words “You really saved the day” or “I couldn’t have done it without you” – indicating a 9-1-1 operator that was instrumental in resolving the incident.
- Identify incidents at risk of potential litigation – e.g. quickly identify complaints so they can be addressed immediately in order to prevent future issues and avoid potential litigation. Automatically identifying calls where some form of the following words were spoken – “screwed up”, “your fault”, “took too long”, etc.
- Source key evidence in support of investigations – e.g. immediately identify key witness statements (from landline and cell phone calls) that correctly identify the suspected perpetrator based on some unique identification such as their name, a specific tattoo, etc.
Audio recordings can offer a mountain of evidence and investigational support when mined correctly for key spoken words. Emergency centers have been recording radio transmissions, witness calls and so forth for a long time now, however, the goldmine of intelligence (audio recordings) sits idle on audio tapes and reels in a back room until needed to address a complaint or support a case. Now, with the advent of highly innovative phonetics-based audio searching software, the spoken word can be leveraged in much the same way text-based content is today.
Without such technology it would be virtually impossible to uncover even similar types of information. Imagine having to manually listen (one by one) to 2,000 tips calls to identify the one that mentions the keywords “yellow scarf”. This would take days. With the new audio searching technology, the call could be discovered automatically, in mere seconds, with no effort at all.
Every 9-1-1 center in the country could potentially benefit from this powerful technology, and once implemented, an emergency center’s capabilities to investigate crime and protect the public would be dramatically improved. What a wonderful value proposition that is!
Tony Procops, founder, President and CEO of CyberTech North America, has over 25 years experience leading recording technology companies. CyberTech is the fastest growing, longest-standing and 3rd largest global call recording provider offering the most secure, open and future-proof solutions for organizations to rapidly improve performance, optimize service, mitigate risk, lower cost and maintain compliance.
For more information, please visit www.CyberTech-NA.com