Commercial intercom, paging, and mass notification design field guide

What these systems are, and the one thing that matters

A commercial intercom, paging, or mass notification system carries the human voice through a building to one of three places: a single door or station, a defined zone, or every occupant at once. That is the whole job described in one sentence, and the equipment that does it ranges from a two-station door box to a campus-wide emergency voice network.

The feature lists are long and mostly beside the point. The measure that decides whether any of it works is speech intelligibility, the degree to which a listener can actually understand the words coming out of the speaker. A paging system nobody can understand is worse than none, because it costs money, it trains people to ignore it, and in an emergency it sends a message that the people who need it cannot use. That last case is where intelligibility stops being an audio quality issue and becomes a life-safety one.

So the real work of designing one of these systems is not picking a touchscreen. It is laying out the speakers for even coverage and intelligibility, choosing between distributed analog audio and IP audio, splitting the building into zones, sizing amplifiers with headroom, and where the system carries emergency messages, tying it to the fire alarm correctly. The cabling that feeds it and the fire alarm it interfaces with each have their own guide; this one is about the audio and the design decisions around it.

Intercom, paging, and mass notification: three jobs, not one box

These three functions get lumped together because they share speakers and wire, but they answer different questions and carry very different consequences. Confuse them and you either overbuild a doorbell or, far worse, treat a life-safety system like a convenience one.

Intercom is point to point. A visitor presses a button at the door, someone inside answers, and the two talk and listen. Paging and public address are one to many: a person at a microphone announces to a zone or to the whole building, and nobody talks back. Mass notification, also called emergency communication, is the one to all that has to work when the building is on fire or a worse event is unfolding. It is the only one of the three that is, by design, life-safety equipment.

Keep the three roles straight from the first design meeting, because the requirements diverge fast. The intercom needs a clear path to a door release. The paging system needs intelligibility and zoning. The emergency system needs all of that plus supervision, survivability, and a defined relationship with the fire alarm under NFPA 72.

Intercom: at the door and station to station

An intercom is a two-way voice link between fixed points. The two most common forms on a commercial job are the entry or door intercom and the room-to-room or station-to-station intercom inside the building.

Entry or door intercom is the one tied to getting people in. A visitor presses the call button, audio (and increasingly video) connects them to a staff station, and the person inside can talk, see, and release the door. On most commercial projects the door intercom is part of the access control design, not a standalone box, so the release contact, the credential reader, and the door hardware all have to agree on how the door releases and re-secures. Get the integration wrong and you either trap people or prop a door that should stay locked.

Room-to-room intercom covers internal communication: a loading dock calling the office, a nurse station to a room, a clean room to a gowning anteroom. The thing to settle early is whether the link is true two-way talk and listen, or push-to-talk where one side releases the button to hear the other. Hands-free two-way is friendlier but harder to keep from feeding back; push-to-talk is simpler and more predictable in noisy spaces. Match the choice to how the people will actually use it, not to the spec sheet.

What is an IP intercom?

An IP intercom is an intercom endpoint that connects over the data network instead of dedicated home-run intercom wiring, usually using SIP, the same session protocol that runs voice over IP phones. Many of these devices are PoE-powered, so a single Cat6 cable carries the audio, the video if it has a camera, and the power. That collapses the cabling and lets the door station register to the same phone or unified communications platform the building already runs.

The move to IP is the direction the field has gone for new work. A SIP door station can ring a desk phone, a softphone, or a mobile app, and a video intercom adds a camera so the person answering can see who is at the door before releasing it. Because the endpoints sit on the network, integrating them with access control, video management, and paging is a matter of configuration rather than running more copper.

The tradeoff is that the intercom now depends on the network and its power. If the switch loses power or the VLAN is misconfigured, the door station goes dark, so PoE budget, switch backup power, and network segmentation become part of the intercom design. Follow the manufacturer's guidance on codecs, SIP registration, and supported PoE class, because interoperability between brands is real but not automatic.

Paging and public address: one voice to many

Paging, or public address, takes one source and pushes it out to many speakers so an announcement reaches a space. Overhead paging through ceiling speakers in an office, horn speakers across a warehouse, a gate call in a transit terminal: all the same idea, a voice announced to people who do not answer back.

Two features define how a paging system gets used day to day. The first is the all-call, the single action that pages every zone at once, used for the building-wide announcement. The second is the ability to combine background music with paging on the same speakers, so the system plays music when nobody is paging and ducks or mutes it the moment someone keys the microphone. That dual use is why so many retail and office systems exist at all; the music pays for speakers that are really there to carry announcements.

The failure mode that defines paging is the one you already know from a big-box store or an old school: an announcement comes over the speakers and it is pure mush. The page reached the space. The information did not. That is the gap between coverage and intelligibility, and closing it is the design work that the rest of this guide keeps coming back to.

Why zone a paging system?

Zoning splits the building so you can page one area without blasting the whole place. A page to the warehouse should not interrupt a meeting in the front office, and a music feed in the cafe should not bleed into the exam rooms. Zones are how a paging system stays useful instead of becoming noise everyone tunes out.

Beyond convenience, zones let the system set priority. A normal page goes to one zone. An all-call overrides everything and goes everywhere. An emergency message, where the system carries one, takes priority over both and seizes the zones it needs. The control equipment enforces that priority order so a routine announcement can never step on an emergency one.

Lay the zones out around how the building is actually occupied and how it has to be evacuated or notified, not around a tidy grid. Tenant boundaries, noise environments, occupancy types, and, for any life-safety function, the notification zones the fire alarm uses should all drive the map. On a mass notification system the zoning often has to match the fire alarm's zones so the two systems give one coherent instruction, which is a point to settle with the fire alarm designer and the AHJ, not to improvise in the field.

What is speech intelligibility, and why does it come first?

Speech intelligibility is the measure of how much of a spoken message a listener can actually understand. It is reported on a scale, most often the Speech Transmission Index (STI) or the related Common Intelligibility Scale (CIS), where higher is better and both run from 0 to 1. It is not the same as loudness. A page can be plenty loud and still be unintelligible, which is exactly what happens in a reverberant gym or a noisy warehouse.

Three things fight intelligibility: reverberation, background noise, and speaker placement. A hard, echoey room smears each syllable into the next so the words overlap and turn to mush. High ambient noise buries the speech under the machinery, the crowd, or the HVAC. And too few speakers spread too far apart leave some listeners far from any source and close to a wall that reflects sound back at them. Adding more speakers, each running quieter and closer to the listeners, usually does more for intelligibility than turning up the ones you have.

This is the design goal that sits above every feature, so design for intelligibility, not just for coverage. Coverage only proves sound reaches a spot. Intelligibility proves the words survive the trip. For any system that carries emergency messages, NFPA 72 makes intelligibility an explicit performance requirement in acoustically distinguishable spaces, and Annex D in the code gives the explanatory background. The specific pass criteria, the measurement method, and whether a quantitative test or a listening test is accepted are set by NFPA 72, the project documents, and the AHJ, so confirm the standard and the target before you commit to a layout.

Speaker layout and density

The speaker layout is where intelligibility is won or lost. Density, the number of speakers and how tightly they are spaced, is the single biggest lever. Higher density means each speaker covers a smaller patch, runs at a lower level, and puts every listener closer to a source, which raises the direct sound relative to the reflected sound and lifts intelligibility. Lower density saves money and gives you fewer speakers running louder, bigger level swings across the room, and worse intelligibility. That is the trade, stated plainly.

For ceiling speakers, a common starting rule is to space them at roughly twice the ceiling height for edge-to-edge coverage, then tighten that spacing in noisy or reverberant rooms. An 8 ft ceiling at that rule puts speakers about 16 ft apart, but treat that as a first pass to refine with the manufacturer's coverage data and a real plan of the space, not a final answer.

Ceiling speakers versus horns is a question of the space. Ceiling speakers give smooth, even coverage in finished rooms with normal ceilings. Horn speakers throw farther and louder and cut through high noise, which is why they own warehouses, gyms, parking structures, and outdoor areas where a ceiling speaker would be lost. Pick the device to the room, and lean on the manufacturer's published coverage angle and SPL data to lay it out rather than eyeballing it.

Coverage: getting above the noise, not just onto the grid

Coverage means even sound across the space with no dead spots, and it means enough level to be heard over whatever noise the room makes. The target most designs work to is a sound pressure level a set margin above the ambient noise. A frequently cited figure for paging is roughly 15 dB above the ambient noise, with 10 to 15 dB often treated as the practical band; confirm the number the project and any life-safety standard call for, since it varies with the application.

The catch is that ambient noise is not one number. A warehouse is loud when the forklifts run and quiet at 6 a.m. Design to the noise level that matters for the message, which for an emergency message is the loud case, because that is when the system has to cut through. Set the target too low and the page disappears under the machinery exactly when it counts.

Even coverage is the other half. The level should not swing wildly as a listener walks the space, dropping into a dead spot under a duct or behind a rack and spiking right under a speaker. Dead spots come from speakers spaced too far apart, obstructions, and rooms whose shape the layout ignored. The fix is density and placement, worked from the manufacturer's coverage data against the real reflected-ceiling plan, not a uniform grid stamped on the drawing.

What is 70V audio?

70V audio, also called constant-voltage or distributed audio (100V in much of the world), is the standard way to wire a large number of speakers across a building from one amplifier. The amplifier puts out a high-voltage line, each speaker has a small transformer that steps that line down, and a tap on that transformer sets how much power that one speaker draws. Higher line voltage means lower current for the same power, which means thinner, cheaper wire and long runs without the loss and impedance headaches of wiring low-impedance speakers in series and parallel.

The tap is the part to understand. Each speaker's transformer offers several taps, commonly labeled in watts, and the installed tap fixes how much of the amplifier's power that speaker pulls and therefore how loud it plays. Tap a speaker higher in a noisy zone and lower in a quiet one, and you balance levels across the building by setting taps rather than by running separate amplifiers.

This is why 70V is the normal choice for paging and PA: many low-level speakers, long distances, and easy level setting per speaker. The design math is to add up the tapped wattage of every speaker on the line and keep that total within the amplifier's rating, which the next section covers. The specific tap labels, the transformer behavior, and any insertion loss are per the manufacturer's data, so size from the actual product, not a generic chart.

Sizing the amplifier

Amplifier sizing on a 70V line starts by summing the tapped wattage of every speaker on that line, then leaving headroom on top. A common rule of thumb is to load the amplifier to no more than about 80 percent of its rated output, leaving roughly 20 percent as headroom, with some designers carrying more for systems that have to handle loud paging and music peaks. Run an amplifier flat out at 100 percent of its rating and you have no margin for the peaks, the sound distorts when it matters, and you shorten the equipment's life.

Headroom is not waste. Speech and music both have peaks well above their average level, and a page keyed by an excited person at the microphone hits harder than a calm test tone. The headroom is what keeps those peaks clean instead of clipped. An amplifier driven into clipping is also a fast way to damage speakers, so the margin protects the whole chain, not just the amplifier.

Match the amplifier sizing to the zones and the priority scheme, because an emergency all-call may need to drive zones that, during normal use, run on separate smaller amplifiers. Size for the worst-case demand the priority logic can create, confirm the load and headroom numbers against the manufacturer's specifications, and account for the power and any backup power the amplifiers draw, especially where the system has an emergency function that has to keep running on secondary power.

Network audio: Dante, AES67, and PoE speakers

Network audio moves the signal as digital packets over standard Ethernet instead of as an analog line on speaker wire. Dante, from Audinate, is the most widely deployed audio-over-IP transport in commercial AV. AES67 is an Audio Engineering Society standard that defines an interoperability layer so different audio-over-IP systems, Dante among them, can exchange streams. Both let you route many channels of audio over the same network infrastructure the rest of the building uses, with the routing done in software.

On the speaker end, IP and PoE speakers take this further: the speaker is a network endpoint that receives its audio and its power over one Cat6 cable. That simplifies the field wiring and makes each speaker individually addressable, so zoning and level setting become configuration rather than physical rewiring. For paging and intercom, the appeal is one converged network carrying audio, control, and power.

The same caution applies as with any networked life-safety-adjacent system: the audio now lives or dies with the network. Switch reliability, PoE budget, network design, latency, and clocking all become audio concerns, and the QoS and VLAN setup is part of the audio design, not an afterthought handed to IT. Follow the manufacturer's architecture and the interoperability profile they support, because Dante, AES67, and proprietary stacks coexist but do not always interchange feature for feature.

What is a mass notification system?

A mass notification system (MNS), also called an emergency communication system (ECS), delivers intelligible voice messages and visible information to occupants during an emergency: fire, weather, a hostile event, a hazmat release, or any situation that needs people to act. It is voice evacuation and more, and it is life-safety equipment, not convenience audio. In NFPA 72, emergency communication systems live in Chapter 24.

Three properties separate an ECS from a paging system. It is supervised, so the system monitors its own wiring and components and reports trouble rather than failing silently. It is survivable, meaning the pathways carrying the emergency message are protected so they keep working long enough to do their job under the conditions of the emergency, with the required survivability level driven by a risk analysis. And it has to be intelligible in each acoustically distinguishable space, because a shelter-in-place or evacuation instruction that no one can understand is the failure that costs lives.

The ECS is also tightly bound to the fire alarm. In many buildings the voice evacuation function is part of the fire alarm system, and where a separate mass notification layer exists, the two are integrated so they give one consistent instruction. The detection, the notification appliances, the survivable circuits, and the acceptance testing that surround this are covered in the fire alarm guide; treat that guide and this one as two halves of the same life-safety conversation. The governing requirements are set by NFPA 72, the manufacturer's listing, and the AHJ, so verify them before you design or price the work.

What is the difference between paging and emergency communication?

Paging is convenience. Emergency communication is life-safety. That single line is the most important distinction in this whole field, and blurring it is how buildings end up with an emergency message running on equipment that was never built to carry one.

A convenience paging system announces lunch, pages a manager, and plays music. If it fails, an announcement gets missed and nobody is harmed. So it is not supervised, not survivable, and not held to a code intelligibility standard. An emergency communication system has to work during the emergency, which is why NFPA 72 holds it to supervision, survivability, and intelligibility, and why its components are listed for the purpose. The two can share speakers in an engineered, listed design, but they are not interchangeable, and a paging amplifier in a closet does not become an ECS because someone wired a microphone to it.

The practical line in the field: if a message is meant to protect people in an emergency, design and install it as an ECS under NFPA 72, listed and accepted by the AHJ. If it is meant to call the stockroom, it can be ordinary paging. Where one system has to do both, the whole system rises to the life-safety requirements for the emergency function. Do not let a convenience budget quietly define a life-safety system; that is the exact decision to put in front of the engineer of record and the AHJ, not to make on the truck.

Fire alarm priority and the emergency path

Where audio carries both routine announcements and emergency messages, the emergency function has priority. A page or a music feed cannot be allowed to step on an emergency message, and the control equipment is built so the emergency input seizes the speakers and overrides whatever was playing. That override and the interface to the fire alarm are part of the design, programmed and tested, not assumed.

The relationship between fire alarm and mass notification is more specific than just fire wins. NFPA 72 requires the fire alarm and emergency communication systems to be integrated, and it allows ECS functions to supersede the fire alarm where the risk analysis calls for it, so a shelter-in-place message can take precedence over a fire evacuation signal when sheltering is the right response. The priority order is set by the design and the risk analysis, not by a default, which is exactly why this gets coordinated with the fire alarm designer and signed off by the AHJ.

Whatever the priority order, the emergency path has to be survivable. The circuits and equipment that carry the emergency message must keep working under the conditions they are meant for, to the survivability level the analysis and NFPA 72 require. An emergency message on a path that the emergency itself can take out is not an emergency system. Verify the survivability requirement, the integration, and the override behavior against NFPA 72, the listing, and the AHJ.

Talkback, two-way, and areas of refuge

Most paging is one direction, but some functions need the message to come back. Talkback lets a paging station listen to a zone as well as announce to it, and true two-way intercom gives both ends a live conversation. The design question is which stations need to hear back and how that audio gets to a staff or security position.

The two-way function that carries code weight is the area of refuge communication system. In many buildings the IBC requires two-way communication at areas of refuge, at certain elevator landings and lobbies, and in some stairways, so a person who cannot self-evacuate can reach help and get an answer. These stations have to provide hands-free two-way communication with a signal confirming the call connected, and they tie to a constantly attended or monitored point such as the fire command center.

Do not satisfy this with an ordinary intercom. Area of refuge systems are life-safety equipment with their own listing and power requirements; recent NFPA 72 editions point to UL 2525 for two-way emergency communication systems for rescue assistance, and they require primary and secondary power with defined standby and operating durations. A standard door intercom does not meet that, so confirm the listing, the power requirements, and the monitoring arrangement with the AHJ before you treat any two-way station as code-compliant for refuge use.

The cabling and the survivable path

These systems ride on a few cable types, and the choice follows the architecture. Distributed 70V paging runs on speaker cable sized for the line and the distance. IP and PoE devices run on structured cabling, the same Cat6 the data systems use, within the channel length and power limits that cabling carries. Intercom may run on either, depending on whether it is analog or SIP. All of it is low-voltage, power-limited work, and the separation, support, firestopping, and grounding for it are covered in the low-voltage cabling guide; follow that guide for the pathway practices rather than reinventing them here.

The cabling decision that is specific to this work is survivability for emergency communication. The circuits that carry an emergency message have to keep working long enough to matter, which is why NFPA 72 sets survivability levels for the pathways, often met with circuit-integrity cable or protected routing rated to keep the path alive through fire for a defined period. The required level comes from the risk analysis and the code, not from habit.

Run the emergency pathways to that survivability requirement and keep them supervised, so a cut or a fault shows up as trouble instead of as a dead speaker discovered during an actual emergency. The specific cable listing, the routing, and the survivability level are set by NFPA 72, the manufacturer, and the AHJ; coordinate them with the fire alarm design, because the two systems often share the same protected pathways.

Sources, priority mixing, and ducking

A commercial audio system usually juggles several sources into the same speakers: a live paging microphone, a background music feed, a doorbell or chime, prerecorded announcements, and the emergency message input. The control equipment decides which one plays when, and that decision is priority mixing.

Priority sets the order. Background music sits at the bottom and yields to everything. A live page outranks music. An emergency message outranks the page. When a higher-priority source keys up, the system either mutes the lower one or ducks it, lowering the music under the page so the announcement is clearly on top, then restoring it when the page ends. A pre-announce chime ahead of a page gives listeners a half second to stop talking and listen, which measurably improves how much of the message they catch.

Set the priority and the ducking so a routine source can never bury an important one, and verify that the emergency input takes absolute priority and seizes the zones it needs regardless of what else is playing. The mixing logic is part of the commissioning test, because a priority scheme that looks right on the configuration screen still has to be proven by keying each source in turn and watching what wins.

How do you commission a paging or ECS system?

Commissioning proves the design in the finished building under real conditions, and it is where most callbacks would have been caught. The functional test walks the whole system: every zone pages, the all-call reaches everywhere, the priority order behaves, music ducks under a page, and the emergency message overrides everything and reaches every required space.

Coverage and intelligibility get tested, not assumed. Walk the space and confirm there are no dead spots and that the level holds above the ambient noise the design targeted. For intelligibility, measure it where the project or the code requires a number, using a method such as an STI-PA measurement with a calibrated analyzer in the acoustically distinguishable spaces, and compare the result to the required threshold. Whether a quantitative measurement or a listening test is accepted, and the exact pass criteria, are set by NFPA 72 and the AHJ for life-safety systems, so confirm the method and the target before test day rather than arguing about it at acceptance.

Test the emergency function as its own event. Trigger it the way it would be triggered for real, confirm it seizes priority, reaches every required zone intelligibly, and that the fire alarm interface and any override behave as programmed. Document every zone, every level reading, every intelligibility result, and the priority tests, because that record is the acceptance package and the baseline the owner's later testing is measured against.

Does code govern intercom and paging systems?

It depends on what the system does. Ordinary convenience paging and intercom are low-voltage installations governed mainly by the NEC for the wiring methods and by the manufacturer's instructions for the equipment. There is no national intelligibility mandate on a system that only announces lunch. The moment a system carries emergency messages, that changes completely.

For emergency communication, NFPA 72 governs, with Chapter 24 covering emergency communication systems and the requirements for intelligibility, supervision, survivability, and the integration with fire alarm. Intelligibility is an explicit performance requirement for these systems in acoustically distinguishable spaces, with Annex D as the explanatory material. Area of refuge two-way systems pull in the IBC and a listing such as UL 2525, and mass notification equipment carries its own product listings. Which editions apply, the exact thresholds, and how they are tested are decisions for the adopted code, the listing, and the AHJ.

The honest summary for the field: design for intelligibility and not just coverage, zone the system and size the amplifiers with headroom, and treat any emergency function as life-safety that is survivable, supervised, and properly tied to the fire alarm. Where the system is life-safety, hedge every intelligibility, survivability, and priority question to NFPA 72, the manufacturer's listing, and the AHJ, and get the design accepted before it is installed, not after.

Maintenance and the annual test

A paging or emergency communication system is not finished at acceptance; somebody has to keep it working. For convenience paging that means periodically confirming the zones still page, the amplifiers still drive their loads, speakers have not been painted over or removed during a remodel, and the music and priority logic still behave.

For emergency communication the testing is not optional and not casual. These systems fall under inspection, testing, and maintenance requirements, and the owner inherits a schedule of periodic functional tests of the emergency function, the supervision, the batteries and secondary power, and the audibility and intelligibility of the message. Run those tests on the cadence the code and the AHJ require and keep the records, because a system that passed at acceptance and was never tested again is a system nobody can vouch for when it has to perform.

The remodel is the quiet killer. Walls move, ceilings change, a noisy new tenant moves in, and the coverage and intelligibility that passed years ago no longer hold. Re-verify coverage after any significant change to a space, especially for a life-safety system, because the building the system was commissioned in is not always the building it has to work in now.

What to document

The record is what lets the next person trust the system without re-deriving it. Capture the zone map and what each zone covers, the speaker layout and model with the tap setting for every device, the amplifier sizing with its load and headroom, the priority and ducking scheme, the intelligibility results by space, and for any emergency function, the survivability level, the supervision, and the fire alarm interface and override behavior.

Tie the as-built to the field, not to a binder nobody opens. Logging the zones, the taps, the test readings, and the acceptance results in a field tool such as FieldOS keeps the record attached to the job, so when a tenant changes, a speaker dies, or the annual test comes due, the layout and the baseline are there to check against instead of lost. The table below is the short version of what belongs in that record.

Common mistakes

• Unintelligible audio that defeats the purpose: loud enough but smeared by reverberation, noise, or too few speakers, so the words do not survive.
• Too few speakers spread too far apart, leaving dead spots and big level swings instead of even coverage.
• No zoning, so every page goes everywhere and people learn to ignore the system.
• Confusing a convenience paging system with an emergency communication system, and carrying emergency messages on equipment never built for them.
• An emergency path that is not survivable or not supervised, so it fails silently or dies in the event it exists for.
• The fire alarm priority and interface never properly programmed or tested, so a routine source can step on an emergency message.
• Never measuring intelligibility, treating coverage as proof the system works when it only proves sound arrives.

Field checklist

Standards and references

For emergency communication, NFPA 72, the National Fire Alarm and Signaling Code, is the governing document. Chapter 24 covers emergency communication systems, including the intelligibility performance requirement in acoustically distinguishable spaces, with Annex D as the explanatory material on speech intelligibility. Survivability levels for the pathways are addressed in the circuits and pathways provisions and applied in Chapter 24 by application. The specific editions, thresholds, and test methods are set by the adopted code and the AHJ, so confirm them before citing a number on a submittal.

Area of refuge two-way communication pulls in the IBC for where it is required and a product listing such as UL 2525 for the equipment, and mass notification components carry their own listings. The wiring methods for all of it fall under the NEC as power-limited, low-voltage work, which the low-voltage cabling guide covers, and the fire alarm side, including detection, notification appliances, and acceptance testing, is in the fire alarm guide.

For the audio design itself, the meaningful references are the audio-over-IP standards and the manufacturer's data. AES67 defines audio-over-IP interoperability, Dante is the common commercial transport, and 70V and 100V constant-voltage distribution is the standard distributed-audio approach. Speaker coverage, tap behavior, amplifier load and headroom, and intelligibility performance all come from the specific product's published specifications. Cite the standard that controls the point, hedge intelligibility, survivability, and the ECS-versus-paging line to NFPA 72, the listing, and the AHJ, and let the project specification override a rule of thumb whenever it is stricter.

Units, terms, and acronyms

These systems mix audio terms, code terms, and network terms, so the same idea shows up under different names across an audio spec, a fire alarm drawing, and an IT handoff. The definitions below are the working ones used through this guide.

Sound level is given in decibels (dB), and design targets are usually stated as a margin above the ambient noise of the space. Intelligibility is reported on the STI or CIS scale from 0 to 1. Distributed audio line voltage is 70V in North America and 100V in much of the rest of the world. Speaker power draw is set by the transformer tap, labeled in watts.

Intercom

A two-way voice link between fixed points, such as a door station and a staff position, or room to room

Paging / PA

Public address: one source announced to many speakers in a zone or building, one direction, no reply

Mass notification / ECS

Emergency communication system: intelligible emergency voice and visible messaging to occupants, life-safety equipment under NFPA 72

Speech intelligibility / STI

How much of a spoken message a listener can understand, reported on the Speech Transmission Index or Common Intelligibility Scale

70V distributed audio

Constant-voltage speaker distribution where transformer taps set each speaker's power, used for many speakers over long runs

Zoning

Dividing the system so a page or message reaches one area without sounding everywhere, and so priority can be enforced

Voice evacuation

A spoken emergency instruction to evacuate or shelter, delivered by the fire alarm or emergency communication system

Survivability

The protection that keeps an emergency circuit working under the conditions of the emergency, to the level NFPA 72 requires

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• What is a low-voltage system?
• What is the difference between conventional and addressable fire alarm systems?
• How many footcandles do you need for an office?
• What is an access control system?
• What is the difference between a feeder and a branch circuit?