When you speak in a room, sound waves travel through the partition walls, glazed screens, doors, and ceiling voids of your office into the spaces next to you. This is not a manufacturing defect or a construction failure, it is physics. The question is not whether sound crosses a boundary, but whether the speech that crosses it can still be understood on the other side.
In a quiet corridor or an adjacent meeting room, a relatively small amount of transmitted sound can be perfectly intelligible. A normal conversation in a meeting room, transmitted at reduced volume through a standard stud partition, frequently remains audible and followable from the space next door, not because the wall is unusually poor, but because the receiving space is quiet enough for that reduced signal to be clearly heard.
This is why privacy is determined not at the wall, but at the listener. The same partition that provides adequate privacy against a noisy reception area may offer almost none against a quiet adjacent office. The variable that matters is the relationship between the level of transmitted speech and the level of ambient sound in the receiving space, the signal-to-noise ratio as experienced by the unintended listener.
Raising the ambient level in the receiving space, through acoustic masking, directly addresses this relationship. It does not change the wall. It changes the acoustic environment on the other side of it, closing the gap between the transmitted speech level and the background until the content becomes unintelligible.
The partition rating tells you only half the story
A partition with a Sound Transmission Class (STC) rating of 38 reduces the sound energy passing through it by a substantial amount. But whether that matters for privacy depends entirely on what is on the other side.
In a receiving space with an ambient noise level of 35 dB(A),a quiet office or corridor, speech transmitted through an STC-38 partition at conversational volume may still be clearly intelligible. In a receiving space with an ambient level of 48 dB(A),raised by masking, the same transmitted speech falls below the threshold of intelligibility.
The partition has not changed. The receiving environment has. This is the mechanism masking uses.
STC, Sound Transmission Class
STC is the single-number rating for how well a partition, door, or glazed screen blocks sound from passing through it. It is weighted towards speech frequencies, making it a reasonable indicator of how much a wall reduces the intelligibility of a conversation on the other side.
The practical benchmarks: at STC 25, normal speech is clearly audible through the partition. At STC 40, loud speech may still be intelligible. At STC 50, most speech becomes inaudible in typical conditions. At STC 60+, near-complete privacy is achievable.
Most standard commercial stud partitions achieve STC 33–42. Most glazed screens and partitions achieve STC 28–35. Neither is sufficient on its own for confidential speech privacy in a quiet receiving space, which describes most meeting rooms and corridors in professional offices.
Rw is the ISO/international equivalent. For building products in the UK, Rw is the standard specification metric.
NRC, Noise Reduction Coefficient
NRC measures how much sound a surface absorbs,not how much it blocks. A ceiling tile with NRC 0.85 absorbs 85% of the sound energy that strikes it within the room, reducing echo and reverberation inside that space. Acoustic foam panels, carpet, and soft furnishings all have high NRC values.
The critical distinction: NRC and STC measure entirely different things. High-NRC materials make the room they are in sound quieter and less echoey. They do not impede sound from crossing the partition into the adjacent space. A meeting room lined with acoustic foam tiles may feel acoustically comfortable to its occupants while transmitting a perfectly intelligible conversation through its walls.
NRC treatment is valuable for reducing echo within a room. It is not a speech privacy measure for adjacent spaces.
Use NRC/α for room acoustics (echo, reverberation); use STC/Rw for partition performance (transmission into adjacent spaces).
STI, Speech Transmission Index
STI is the metric that determines whether a conversation is actually private. It measures speech intelligibility in a receiving space on a scale from 0 (completely unintelligible) to 1 (perfectly clear), taking into account the level of transmitted speech, the ambient noise level in the receiving space, and any reverberation present.
Research by Dr. Valtteri Hongisto at Turku University of Applied Sciences established that when the STI in a receiving space drops below 0.20, speech ceases to be intelligible, it is heard as an undifferentiated ambient sound rather than meaningful communication. This is the target that every masking installation is calibrated to achieve.
STC tells you about the wall. STI tells you about the listener’s experience on the other side of it. The two are related, but masking allows STI to be driven below the intelligibility threshold regardless of what the wall’s STC rating is.
Hongisto, V. (2005). Indoor Air, 15(6), 458–468. IEC 60268-16 defines the STI measurement methodology.
What the masking signal sounds like
The masking signal is not white noise. It is a spectrally shaped signal, engineered to match the frequency profile of human speech, that blends into the acoustic environment of the receiving space within minutes of exposure. Most people passing through a masked corridor or waiting area do not consciously perceive it. The effect is experienced as a slightly fuller ambient sound, not as an intrusive noise. Within a day of installation, most occupants stop registering it entirely.
The STI intelligibility scale
The Speech Transmission Index (STI) determines whether a conversation is truly private in the receiving space. A masking installation targets an STI below 0.20 in every zone where privacy is required.
STI in receiving space
What this means for privacy
STI in receiving space: 0.76 – 1.00
What this means for privacy: Conversation clearly audible and followable, no privacy
STI in receiving space: 0.60 – 0.75
STI in receiving space: 0.45 – 0.60
STI in receiving space: 0.30 – 0.45
STI in receiving space: Below 0.20
Unintelligible, confidential privacy achieved
Hongisto, V. (2005). Indoor Air, 15(6), 458–468. STI methodology per IEC 60268-16 and ISO 9921. The STI <0.20 “confidential” threshold follows ANSI/ASA S12.70-2016 (R2025), a US healthcare-facilities standard.
Masking is applied where the listener is
The principle is the same wherever speech needs to be made unintelligible: the masker sits in the space where the listener is, not the space where the speaker is. The listener might be in a corridor outside a meeting room, in a waiting area next to a consultation room, or in an adjacent meeting room overhearing the room next door.
Where two rooms can overhear each other through a shared partition, each room is simultaneously a speaking space and a listening space, so maskers are installed in both. The general rule holds: put masking wherever a listener might otherwise pick up intelligible speech from elsewhere.
Confidential
Normal
Marginal
Poor
For professional and regulated businesses, financial services, legal, HR, healthcare, the relevant target is Confidential Privacy. This means that a person immediately outside your meeting room, in the corridor beyond your office door, or seated in your waiting area cannot understand what is being said on the other side of the partition, even if they can detect that a conversation is taking place.
Confidential Privacy is defined as a Privacy Index (PI) of 95 or above, meaning less than 5% of sentence content is intelligible to an adjacent listener. This is not a subjective judgement. It is a measurable outcome, verified after installation using calibrated test equipment and recorded in a written commissioning report.
Most existing commercial offices, with standard stud or glazed partitions and quiet corridors, fall in the Marginal or Poor category without masking. With correctly installed and calibrated masking in the receiving zones, Confidential Privacy is achievable in the same spaces without any change to the partition.
ANSI/ASA S12.70-2016 (R2025)
Criteria for Evaluating Speech Privacy in Healthcare Facilities
This is a US standard, scoped to healthcare facilities, where demonstrable speech privacy supports HIPAA compliance. Its four-level privacy classification is widely used as a reference in professional offices, and we apply that logic alongside the UK and European standards opposite, which are the primary framework for our market.
It defines a privacy classification system and the supporting metrics, STI, Speech Intelligibility Index (SII), Articulation Index (AI), and Privacy Index (PI), used to evaluate installed systems.
American National Standards Institute / Acoustical Society of America. Current version: ANSI/ASA S12.70-2016 (reaffirmed 2025). www.ansi.org
The UK and European framework
For the UK and European market, speech privacy in workplaces is addressed by modern international standards adopted as British Standards:
→ BS EN ISO 3382-3:2022 – measurement of room acoustic parameters in open-plan offices (distraction distance, comfort distance, spatial decay of speech).
→ BS ISO 22955:2021 – acoustic quality of open office spaces; sets design targets.
→ IEC 60268-16 and ISO 9921 – definition and assessment of the Speech Transmission Index (STI).
→ BS 8233:2014 – UK guidance on sound insulation and noise reduction for buildings.
These are the current, UK/European-aligned reference standards; the privacy classes above are a useful US-origin overlay.
Standards text is published by ISO/BSI/IEC. Specific target figures are validated against the source standards before reliance.
ASTM E1573,Standard Test Method for Measurement and Reporting of Masking Sound Levels Using A-Weighted and One-Third-Octave-Band Sound Pressure Levels (current edition E1573-22),provides the standardised methodology for measuring and reporting the sound levels of installed masking systems. It is a measurement and reporting method: ASTM’s own scope states it is intended only to measure and report masking sound levels and does not evaluate the overall acoustical environment. The same method is applied to enclosed rooms, corridors, and any space in which a masking system is installed.
We use E1573 to measure and report three parameters; the target values themselves are industry design conventions, not levels mandated by the standard:
→ Level: The overall A-weighted sound pressure level of the masking signal, which we design to a typical 45–48 dB(A),sufficient to achieve masking without being intrusive.
→ Spectrum: The frequency content of the masking signal, shaped towards speech frequencies, verified through one-third octave band measurements.
→ Uniformity: The masking level across the zone, which we hold within ±2 dB, so there are no areas of insufficient masking or intrusive over-level.
Every installation we commission concludes with measurements taken in all treated zones and a written commissioning report documenting the results. This report confirms which privacy classification has been achieved in each space, the corridor, the adjacent office, the waiting area, and provides the documented evidence of a specific technical control.
ASTM International. www.astm.org
The commissioning report as compliance evidence
For regulated organisations, the commissioning report is more than a technical record. It is the document that answers the question: “What specific, measurable step did you take to protect confidential spoken information?”
A report that states, room by room and zone by zone, that the measured STI in the receiving space falls below 0.20 and that Confidential Privacy has been achieved is a demonstrably reasonable technical precaution, supporting the kind of “appropriate technical and organisational measures to ensure a level of security appropriate to the risk” required by UK GDPR Article 32, the FCA’s general systems and controls expectations (SYSC), and the SRA’s confidentiality requirements.
This is the difference between having addressed the problem and being able to prove you have addressed it.
Information only, not legal advice; confirm interpretation with your legal team.
Dr. Valtteri Hongisto
Turku University of Applied Sciences, Finland. Established that speech intelligibility, not overall noise level, is the operative variable for privacy and cognitive distraction. Defined the STI threshold of 0.20 below which speech becomes unintelligible. Validated masking as an effective intervention across enclosed and open environments.
Indoor Air (2005) · Applied Acoustics (2016) · Journal of the Acoustical Society of America
Dr. John S. Bradley / NRC Canada
National Research Council Canada, Acoustics Laboratory. Developed the objective speech privacy metrics, including the Privacy Index, used in ANSI/ASA S12.70. Validated the effectiveness of masking in achieving Confidential Privacy without structural change to partitions.
Canadian Acoustics, 31(2), 2003 · ICA Proceedings, 2004 · NRC Construction Technology Updates
Cornell University
Human Factors and Ergonomics Laboratory. Documented that intelligible speech, whether originating in the same space or transmitted from an adjacent one, impairs concentration, elevates stress markers, and reduces task performance. The practical implication applies equally to open floors and partitioned environments: intelligibility, wherever it originates, is the variable to control.
Workplace environment research, Cornell University
Eindhoven University of Technology
Study on adaptive masking systems. Found that systems which adjust in real time to changing ambient conditions outperform static systems in achieving consistent privacy performance throughout the day. Relevant to multi-room environments where traffic patterns vary.
Eindhoven University of Technology, adaptive masking study
Need to understand the regulatory obligations?
Alternatively, get a budget estimate or request a site survey.
