Voice Evacuation and Speech Intelligibility for the NICET Exam: ADS, STI Targets and Testing (NFPA 72, 2026)
NICET exam guide to voice evacuation: audibility vs. intelligibility, acoustically distinguishable spaces, STI and CIS targets, and how systems are tested.
Problem this solves
This guide solves the next-step problem for Fire Alarm candidates: it explains what matters, then gives you a direct way to test that knowledge with practice questions instead of guessing what to study next.
TL;DR
A voice evacuation system is only useful if people can understand the words, not just hear that something is talking. NFPA 72 handles this with two ideas the NICET exam likes to test: the acoustically distinguishable space (ADS), which is the unit you design and test in, and speech intelligibility, which is measured with a score between 0 and 1. A widely cited acceptance target is an STI of at least 0.45 at 90 percent of measurement locations in an ADS, with an average of at least 0.50. Not every space needs measured intelligibility, but every space with voice notification still needs to be audible. Learn the concepts and the logic, then confirm exact paragraph numbers against the edition your exam is built on.
Try it now — no account, no download
Practice Fire Alarm questions free
Answer 5 Fire Alarm practice questions with instant explanations before you keep reading.
Fire Alarm · Question 1 of 5
What is the primary difference between a conventional and an addressable fire alarm system?
Pick an answer to see the explanation + an instant AI breakdown.
What the NICET Exam Is Built On
NICET's Fire Alarm Systems page lists Level I at 85 questions in 110 minutes and Level II at 110 questions in 155 minutes, delivered by computer at Pearson VUE. The same page says the current exams are based on the 2022 edition of NFPA 72, with a later edition expected on a date still to be determined. Fees listed there, effective August 1, 2026, are $235 for Level I and $325 for Level II. Verify all of this at nicet.org/fas before you book, because NICET can change it. Why edition matters here: NFPA renumbered parts of the notification chapters between editions, and published articles on this topic cite different paragraph numbers for the same idea. This post therefore teaches the concepts and leaves paragraph numbers for you to confirm in your own copy of the code.
Free: the 20 facts Fire Alarm exams keep testing
Top 20 most-tested facts for your exam. Delivered instantly.
Voice Evacuation vs. Tones: Why Intelligibility Exists
Horns and bells tell people something is wrong. A voice evacuation system, one type of emergency communications system, tells them what to do: which stair to use, whether to move or stay put, whether the event is on their floor. High-rises, hospitals, assembly spaces, schools and airports commonly use voice because evacuating everyone at once is not always the right answer. The catch is that a speaker can be plenty loud and still be useless. Hard surfaces, echo, background noise and poor speaker placement smear words together. Intelligibility is the code's way of asking whether a listener can actually understand the message. It is a separate question from audibility. See our guide to [audibility, public mode and private mode](/blog/nicet-fire-alarm-audibility-public-mode-private-mode-sound-pressure-nfpa-72-2026) for the loudness side, and the overview of [emergency communications and mass notification](/blog/nicet-fire-alarm-emergency-communications-systems-mass-notification-nfpa-72-chapter-24-2026) for the system context.
The Acoustically Distinguishable Space (ADS)
You cannot test a whole building as one blob of sound, so NFPA 72 asks the designer to divide it into acoustically distinguishable spaces. An ADS is a space that differs from its neighbors because of its acoustics, its environment or how it is used, for example in reverberation time and ambient noise level. A carpeted office, a stone-floored lobby and a loud mechanical room are different acoustic problems, so they are different ADSs. Two exam points follow. First, the designer identifies the ADSs during planning, and the boundaries are shown on the drawings. Second, the code does not require measured intelligibility in every ADS. Some spaces, such as mechanical rooms, may be exempt from the intelligibility requirement, but where voice is used they still need the signal to be audible. The reasoning the exam rewards: intelligibility is required where occupants need to understand instructions, and the annex and AHJ guide the exceptions.
Measuring It: STI, CIS and the Numbers
Intelligibility is usually reported on one of two scales. The Speech Transmission Index (STI) runs from 0 to 1. The Common Intelligibility Scale (CIS) is a related 0 to 1 scale, and published conversions put CIS 0.65 near STI 0.45 and CIS 0.70 near STI 0.50. An SDM article that walks through the 2010 edition's annex describes acceptable intelligibility as at least 90 percent of measurement locations within each ADS reaching STI 0.45 (CIS 0.65), with the average across the ADS at least STI 0.50 (CIS 0.70). Other industry write-ups for current editions repeat the same thresholds. Treat the two-part shape as the exam pattern: a floor that nearly all points must meet and a higher average for the space. In the field, the technician plays a test signal through the voice system, commonly modulated pink noise, and reads the score at each measurement point with a handheld STI or STIPA meter. Points are spread through the space at ear height. For a prerecorded message the signal goes through the panel; for a live microphone, a speaker-style test device replays the signal into it. Subject-based methods, where trained listeners score words, also exist, but instrument testing is the usual route. Intelligibility can also be predicted at the design stage, which is why layout choices matter. More speakers at lower volume usually beat a few loud speakers in a reverberant room.
Common Mistakes on Exam Day
The first mistake is mixing up audibility and intelligibility. Audibility is about sound level relative to ambient noise. Intelligibility is about whether words can be understood. A question about a loud but echoing atrium is almost always testing intelligibility. The second is assuming every space must be measured. If the stem describes a mechanical room or a space with no occupant instructions needed, expect that the answer involves audibility or an exemption rather than a full STI survey. The third is scrambling the numbers. Remember the shape: a high share of points (90 percent) clearing a lower floor, plus a higher average. If an answer choice has them reversed, it is wrong. The fourth is forgetting that a voice system is tested more than once. New systems are verified at acceptance, then checked on a recurring schedule; see the [acceptance testing guide](/blog/fire-alarm-acceptance-testing-nfpa-72-chapter-14-nicet-exam-2026) and the [ITM frequency schedule](/blog/nicet-fire-alarm-testing-frequencies-nfpa-72-itm-schedule-2026). Details of what must be re-measured can vary by edition and by AHJ.
Study Plan
Spend one session on the vocabulary: ADS, audibility, intelligibility, STI, CIS, STIPA. Spend a second building a one-page comparison of tone signaling and voice signaling, including where each fits. Spend a third on scenario questions: a reverberant lobby, a noisy mechanical room, a hospital corridor, a school gym. Then review how voice fits with the rest of the notification chapter using the [notification appliance guide](/blog/nicet-fire-alarm-notification-appliances-candela-strobe-spacing-nfpa-72-chapter-18-2026). VoltExam's Fire Alarm Prep app has 702 supplemental Fire Alarm Systems questions for Level I and II topic review. Practice results do not predict an official result and are not a substitute for NICET's own requirements. [Download the Fire Alarm Prep app](/apps/fire-alarm) and drill notification questions on your breaks, or try free NICET practice questions on VoltExam first. [LINK: /apps/fire-alarm]
Sources
NICET, Fire Alarm Systems certification program page (exam questions, time, fees, NFPA 72 edition): https://nicet.org/fas SDM Magazine, Measuring voice intelligibility for emergency communications systems (2014, discusses the 2010 edition of NFPA 72, Annex D thresholds and the ADS concept): https://sdmmag.com/articles/90530-measuring-voice-intelligibility-for-emergency-communications-systems QRFS blog, EVACS speech transmission index tests and equipment (2026, industry summary; paragraph numbers differ from the SDM article, so confirm against your edition): https://blog.qrfs.com/501-evacs-speech-transmission-index-tests-and-equipment/ NFPA 72, National Fire Alarm and Signaling Code, Chapters 18 and 24 and Annex D (paragraph numbers intentionally not quoted; confirm against your edition): https://www.nfpa.org/codes-and-standards/nfpa-72-standard-development/72
Studying with VoltAI
Stuck on a Fire Alarm question? Ask VoltAI why.
VoltAI is a built-in AI tutor that explains every practice answer in plain language — citing the actual code, protocol, or standard — and answers your follow-ups until it clicks. Most prep apps just tell you the right letter; VoltAI teaches the reasoning the exam actually tests, so you pass by understanding, not memorizing. Free to try, no account.
Try VoltAI free →Your next best step
Turn this guide into practice
Use the article to understand the topic, then do a short web practice session to find your weak spots. Paid web access is optional after the free preview.