The combination of audio analytics technology with STENTOFON intercom systems provides security personnel with a powerful tool to be able to act fast or even pre-emptively when a threat is detected. The combination allows STENTOFON intercoms and information/help points to automatically alert guards to car thefts or break-ins (car alarm detection), office or building break-ins (glass break detection), verbal and physical violence in hospitals, prisons, reception areas, (aggression detection), and firearm crimes (gunshot detection). These analytics will do what video can’t: detect crimes and events by analysing audio; detect off-camera events or areas where cameras are not located; and detect events regardless of lighting conditions.

The analytics technology is designed to recognize complex sounds by using advanced psychoacoustic modelling, the science behind how humans distinguish and understand the meaning of sound. The CoreLogger™ sound recognition software can detect a variety of sounds depending on the type of Sound Packs™ installed, even with high levels of ambient noise.

The STENTOFON analytics technology is licensed to Zenitel by Audio Analytic, a UK based company which specializes in sound detection and classification. The information is this whitepaper is provided in cooperation between Zenitel and Audio Analytic.


The different analytics

Different sounds have different characteristics. Detecting these characteristics is the basis on which the analytics work; listening out for the specific characteristics that define the sound. These characteristics do not change, although the sound itself may sound differently to the human ear as this depends on the acoustical properties of the environment.


The characteristics of each type of sound are captured in specific Sound Packs™.


Breaking glass

The glass break Sound Pack has been designed to allow detection of standard window glass being broken. Breakage of a range of different glass types can be detected, including laminate, plate, wired and tempered glass, as will be commonly found in commercial and residential buildings. Breaking of glass is characterized by the initial sound of the window being broken, the impact, followed by the glass actually shattering.

Breakage of different thicknesses and sizes of glass have been characterized in the Sound Pack, as well as a range of different “implements” used to break the glass.

The Sound Pack is capable of detecting the following thicknesses and sizes of glass:

▪ Plate: 2.4 mm – 10 mm

▪ Tempered: 3.2 mm – 10 mm

▪ Wired: 6 mm (thickness wired glass is typically supplied in)

▪ Laminated: 4.4 mm - 10.8 mm

▪Minimum glass size, all types of glass 300 x 300 mm

The system can detect up to 10 meters from sound source using the microphone of STENTOFON IP-intercom stations. Note in the case of laminated glass protection is only provided if both panes of glass are broken and all types of glass must be mounted in a solid support such as a wall or door.

By using the microphone of a STENTOFON IP intercom station as the detector instead of dedicated glass break detectors there is an added advantage of there being a loudspeaker at that location. When the breaking glass triggers an alarm a guard is now able to warn off the intruders which will in many cases mean that the final damage will be limited to the broken window pane. When used in this way the glass break detection is a true perimeter protection mechanism as, unlike a PIR or motion sensor, it detects intrusion while the person is still outside the property.

This analytic is applicable to any end user application where there is a need to detect break-ins, including retail, commercial offices, hospitals, etc.


Car alarms

The car alarm Sound Pack has been designed to allow detection of the seven standard types of car alarms used by major car manufacturers.

The seven major car alarms are characterized by looping progressions of specific tonal patterns.

It can operate up to 50 meters from the sound source using the microphone built into a STENTOFON IP intercom station

In any location with a large number of parked cars, such as a car dealership’s forecourt or a poorly lit office or hospital car park, there can be a problem that traditional video surveillance might not detect incidents of car break-in or theft, or conversely, motion detection using video analytics generates false alerts as a result of people moving quite innocently around the vehicles.

Use of audio analytics in such circumstances can greatly enhance system functionality, allowing security staff to be alerted to car break-ins at times when their cameras cannot detect the incidents.

This analytic is applicable to any location where there is a high density of parked cars: for example transportation hubs, stations, airports, car dealership forecourts, shopping malls, offices, hospital or college car parks, etc.



The aggression Sound Pack has been designed to allow detection of aggressive speech or shouting, symptomatic of a developing security incident where staff, customer or resident safety is likely to be compromised.

Aggression is characterized by the specific pitch, tone and intonation changes that occur in the voice patterns in response to someone getting aggressive.

It can operate up to 10 meters from sound source using the microphone of STENTOFON IP intercom stations.

Use of the aggression analytic allows security staff to be alerted to an aggressive incident that video only systems are unlikely to spot. This allows security staff to be alerted to aggressive incidents early, before they develop into actual physical aggression, allowing security staff to provide early intervention and improve staff (or customer) protection and safety. Any location where there is regular interaction between people and where there is the possibility of aggressive behaviour can benefit from this analytics capability, for instance a hospital accident and emergency reception desk, retail customer services desk, late night convenience stores or petrol stations, or, e.g. colleges, city centres, prisons, police custody centres, etc.

In addition, detection of aggressive speech/shouting can be used to trigger pre/post audio recording of the incident which would be impossible using video only security systems.

Legislation that prohibits audio recording in public spaces normally allows recording of audio if a security incident has been detected, and hence the audio analytic can provide significant value for staff/customer protection and offender prosecution, without compromising privacy issues. You are advised to check on specific legislation in your region to ensure that recording audio under such circumstances is permitted.



The gunshot Sound Pack has been designed to allow detection of a range of firearms being discharged, covering the types of guns most commonly used in civilian gun crimes.

Gunshots are characterized by the unique muzzle blast associated with a range of weapons typically used in civilian gun crimes in the Americas, Europe, Middle East and Australasia.

The types of weapons are

▪ Hand guns (including 9mm automatics and revolvers with and without muzzle diffusers)

▪ Shot guns (including 20-gauge, .410 and 12 Bore)

▪ Riffles (.22 and 7.62 mm bolt action)

▪ Automatic rifles (including AK-47, AR-15 and Uzi submachine gun)

Note that this Sound Pack detects the muzzle blast so protection is only provided as long as the weapon is not silenced which is not normally the case with civilian gun crime.

It can operate up to 100-200 meters from the sound source using the microphone built into a STENTOFON IP intercom station. Incidents of gun crime are increasing throughout the world and systems to detect gunshots and alert security staff is already being deployed, although the majority of these systems are high cost with significant additional installed infrastructure required.

Use of the analytic integrated with the STENTOFON intercom system allows gunshots to be detected by for instance the help points on campuses or throughout city centres providing dual use for such help points. Also intercom stations within premises can be used for the detection of gunshots. This allows security staff to activate the most appropriate emergency evacuation procedures and also direct police to the relevant area of the site/building.

This analytic is directly applicable to college/university campuses, hospital accident and emergency reception areas, late night protection of lone workers in convenience stores, motels or petrol stations, or city centres.

Share this page...

© Copyright 2018 STENTOFON Communications Australia | Website Design by Sites n Stores