A technique that adds a controlled, low-level shaped noise to a space so that overheard speech becomes unintelligible. Functionally identical to sound masking; the two terms are used interchangeably on this site.
An older but still useful metric for speech intelligibility, expressed on a 0 to 1 scale. AI is derived from the signal-to-noise ratio in each speech frequency band. It has largely been superseded by STI in modern measurement work but remains common in privacy-focused literature.
American standard for measuring speech privacy in open offices using the Articulation Class. Describes a method for quantifying how intelligible speech is between adjacent open-plan workstations.
American standard for measuring speech privacy between enclosed rooms using the Speech Privacy Class. The closest analogue to STI measurement focused on closed-room confidentiality.
American test method (current edition E1573-22, “Standard Test Method for Measurement and Reporting of Masking Sound Levels Using A-Weighted and One-Third-Octave-Band Sound Pressure Levels”) covering how a masking sound level is measured and reported. It is a measurement-and-reporting method only and does not set or mandate a target level; the commonly cited 45–48 dB(A) figure is an industry design convention, not an E1573 requirement.
UK guidance standard, “Guidance on sound insulation and noise reduction for buildings”. It gives recommended internal noise levels and design guidance for sound insulation and noise control in occupied buildings, including offices.
The UK/European measurement standard for the room acoustics of open-plan offices. It defines how to measure single-number parameters such as the distraction distance, the comfort distance and the spatial decay rate of speech (D2,S).
Standard on the acoustic quality of open office spaces. It sets out design targets and approaches for different ways open-plan spaces are used, and works alongside BS EN ISO 3382-3 for measurement.
An open-plan room-acoustic metric defined in BS EN ISO 3382-3: the distance from a speaker at which the A-weighted sound pressure level of speech falls below a defined comfort threshold. A shorter comfort distance indicates speech is contained more locally.
A unit of sound level weighted to match the frequency response of the human ear at typical conversational levels. Most measurements relevant to office speech privacy are reported in dB(A); a typical office background sits around 38–42 dB(A) and a calibrated masking system raises this to about 45–48 dB(A).
An open-plan room-acoustic metric defined in BS EN ISO 3382-3: the distance from a speaker at which the Speech Transmission Index falls below 0.50, i.e. the distance beyond which speech becomes hard to follow and is therefore less distracting. A shorter distraction distance indicates a better-controlled open-plan space.
An open-plan room-acoustic metric defined in BS EN ISO 3382-3: the rate, in decibels, at which the A-weighted speech level falls for each doubling of distance from the speaker. A higher D2,S means speech decays more quickly with distance, which generally supports better speech privacy.
The international standard that defines the Speech Transmission Index and sets out the objective measurement method for it. The primary reference document for any acoustic survey that needs to produce a defensible STI figure.
International standard on the assessment of speech communication, which uses the Speech Transmission Index to rate intelligibility against descriptive categories (from “bad” to “excellent”). Read alongside IEC 60268-16, which defines how STI is measured.
A small speaker optimised to deliver the masking signal. The two terms are used interchangeably on this site: “masker” emphasises the function (the unit that produces the masking signal), “speaker” describes the underlying hardware. Three common types in offices: plenum, tile, and on-ceiling.
A single-number rating of how absorptive a surface is across the speech frequency range, between 0 (fully reflective) and 1 (fully absorptive). Suspended ceiling tiles typically have an NRC of 0.55 to 0.85; a hard plasterboard ceiling has an NRC near 0.05.
A masker mounted directly on the underside of the ceiling, used when there is no plenum void above the room (for example, in a slab-soffit office). On-ceiling maskers are visible in the room but require no void access; they fix to the underside of the ceiling itself.
A noise signal whose energy is inversely proportional to frequency, equal power per octave. Pink noise sits closer to the natural balance of speech and ambient sound than white noise, so a pink-noise-derived masking signal is less audibly intrusive.
The space between a suspended ceiling and the structural slab above. In most commercial offices this void is between 200 mm and 600 mm deep and is the natural home for plenum maskers.
A masker designed to be installed in the ceiling plenum, projecting sound downward through the suspended ceiling. The most common masker type in UK offices because it is invisible from the room and does not require ceiling tile modification.
The complement of Articulation Index: PI = (1 − AI) × 100, expressed as a percentage. A PI of 80 corresponds to “normal” privacy; 95 corresponds to “confidential” privacy.
The time, in seconds, for a sound to decay by 60 dB after the source has stopped. Long RT60 values mean a “live” room with lots of reflections; short values mean a “dead” acoustically absorptive room. Office RT60 is typically 0.3 to 0.6 seconds.
A technique that adds a controlled, low-level shaped noise to a space so that overheard speech becomes unintelligible. The masking signal is shaped across the speech frequency range, roughly 300 Hz to 4 kHz, where it is most effective at covering intelligible speech. Identical in meaning to acoustic masking.
Reducing sound transmission between spaces using construction: added mass, isolation, decoupling, sealing, and damping. A different intervention from sound masking, with different costs and a different outcome (see Soundproofing vs Masking).
How much of a spoken message a listener can correctly understand. Measured by STI, AI, or speech-reception tests using calibrated word lists.
The condition of a space in which spoken conversation cannot be understood by an unintended listener outside the conversation. The end objective of every system on this site.
A single-number rating of a partition or door’s airborne sound reduction across speech frequencies. Standard office partitions rate STC 35 to 40; properly sealed full-height partitions rate STC 45 to 50.
The modern standardised metric of speech intelligibility, on a 0 to 1 scale defined by IEC 60268-16 and ISO 9921. STI < 0.20 is the threshold most speech privacy installations target on the listener side of a partition, the level at which speech ceases to be intelligible. This “Confidential” threshold follows ANSI/ASA S12.70-2016 (R2025), a US healthcare-scoped standard.
A masker that adheres to the upper surface of a ceiling tile, used in offices with shallow plenum voids that cannot accommodate a plenum masker.
The unit that generates the masking signal and drives one zone of maskers (speakers); typically up to eight maskers per controller. Each controller needs a standard mains socket; one controller per acoustic zone.
