Maybe audio communication was actually ahead of its time, because today, this technology can enhance an integrated system specification with increased situational awareness, real-time data and an accurate assessment of current security or safety risks and threats.
Audio detection is a decades-old security technology. Robert Baxter of Anderson, Ind., installed microphones within the walls of a building to listen for termites and pinpoint their location. He also invented other sound-related devices.
In 1960, Baxter incorporated Sonitrol—derived from “sound” and “control”—to manufacture and market his innovation as verified electronic security.
Audio technology has advanced significantly from its initial days as a “listen in” device designed to detect structural break-ins, smashing glass or other acoustic variations. Audio products today consist of intercoms, loudspeakers, microphones, gunshot detection and advanced multisensor audio analytics. The technology is also deployed in emergency call boxes and paired with strobes, cameras, access control and other intrusion detection devices.
Artificial intelligence (A.I.) and machine learning algorithms have moved the product category into integrated, intelligent communications. A.I. has pushed the popularity and limits of audio analytics, with advanced voice control that leverages natural language processing. It is powered by networked and IP-enabled microphones, speakers and audio processing and control software, according to the Security Industry Association (SIA).
“Today, transitioning from analog to digital audio solutions is much easier. One reason is that the network has become the backbone for data capture, communications and security. The second reason is that most IP-networked audio solutions are software-based, making them much more flexible and scalable,” according to a guide published by SIA’s Audio and Intelligent Communications Working Group, “Safe and Sound: A Primer on Audio and Intelligent Communications Applications for Security.”
The most recent innovation in audio analytics, according to the SIA guide, is the introduction of multisensor awareness. This means that multiple sensors working together help determine whether the sound heard was a gunshot or false positive. In the future, we can expect that combining audio and video metadata with other data sources (weather, social media, etc.) will lead to better trend analysis and predictive systems that can make autonomous decisions.
The power of hearing
The first uses of audio analytics focused on listening for an increase in decibel level. If it passed a preset threshold, an alarm sounded. False positives and alarms were a problem, because it was difficult to distinguish between sounds. Analytics solved those problems, and new developments leveraging A.I. analyze various frequencies and patterns that identify dangerous or violent situations or threats.
The move to digital audio removes what commonly plagued analog audio: interference from other electrical equipment, communication crosstalk and signal degradation over long cable runs. Another benefit of IP audio, like other network-based technologies, is less cabling and lower installation costs. Audio streams can be programmed through software, and multichannel audio can cover long distances with near-zero latency and no signal loss.
With the technology now network- and software-based, audio integrates readily with other products for holistic situational awareness. Monitors can listen to live and recorded audio, resulting in detailed information on the threat level so first responders are better prepared. Two-way communications allow users to voice-down suspected intruders or direct occupants to safe areas. Audio analytics now measure a host of audible changes in the environment, such as verbal aggression and screams, fire alarms, coughing or other personal emergencies, car sirens, medical alarms, gunshots and even defective machinery/condition monitoring.
Audio verifies alarm status
According to the U.S. Department of Justice, more than 90% of emergency alarms are unnecessary and pose a serious threat to police department effectiveness and public safety. Verified alarm response from audio technologies reduces false dispatches and increases apprehensions, according to the Partnership for Priority Verified Alarm Response. The organization has produced a document, “Audio Verified Alarms Best Practices,” which provides insights into properly leveraging audio for alarm verification.
Audio analytics has also moved into edge and cloud options, with the processing power in cameras, intercoms and standalone audio devices transforming them into smart detection products rather than simply static ones capturing data and information later.
Remember to check applicable laws regarding microphones, audio and privacy in your area before starting work. Aside from signage, as an industry best practice, audio systems should not record sound continuously. Follow other recommendations from manufacturers for best results.