Unified Metric for Speech Privacy in Open and Closed Office Environments

Summary of a paper presented at the joint 45th AIVC conference and ASHRAE 2025 IEQ conference “IEQ 2025: “Rising to New Challenges: Connecting IEQ to a Sustainable Future” will be held on September 24-26, 2025, in Montreal, Quebec together with the 13th TightVent and the 11th venticool conferences.

Key words: Speech Privacy; Indoor Environmental Quality (IEQ); Open-plan Offices; Closed Offices; Speech Privacy Class (SPC); Privacy Index (PI); Speech Privacy Potential (SPP)

Introduction

Speech privacy is a crucial component of Indoor Environmental Quality (IEQ) in office environments, directly impacting occupant satisfaction, productivity, and overall well-being. Despite decades of research and the development of multiple objective metrics, speech privacy remains one of the lowest-rated aspects in post-occupancy evaluations of office buildings [1], [2], [3]. A key challenge is the lack of consensus on which objective metric best represents occupants’ subjective experiences, particularly across different types of offices. The study focuses on the three most widely applied metrics in North America: Speech Privacy Class (SPC, ASTM E2638 [4]), Privacy Index (PI, ASTM E1130 [5]), and Speech Privacy Potential (SPP, unstandardized). Few studies have correlated these metrics with subjective responses collected in the field [2], [3], [6], [7], [8], [9], [10], and none have evaluated them across both open and closed offices using the same dataset. By integrating occupant surveys with in-situ acoustic measurements, this study aims to bridge that gap and provide practical guidance for designers, acousticians, and standards developers seeking a single, unified metric capable of reliably predicting occupant perceptions of speech audibility, intelligibility, and distraction in both open-plan and closed office settings.

Methodology

Subjective Surveys

Subjective data was collected through an online survey administered to 82 office workers across three office locations in Quebec, Canada: a university facilities management office and two engineering offices. Participants represented a range of professional roles and worked in diverse office configurations, including open-plan workstations, private offices, shared offices, and meeting rooms.

Respondents were asked to assess their perceptions of speech privacy relative to their primary workstations and commonly used adjacent spaces. They evaluated unwanted speech in terms of audibility (whether speech could be heard), intelligibility (whether speech could be understood), and distraction (whether speech interfered with work tasks) from sources for which they were not the intended recipient. Participants rated both their experienced and expected speech privacy between named office locations, allowing comparison between perceived deficiencies and desired performance.

Objective Acoustic Measurements

Objective measurements were conducted in the same office environments evaluated by survey participants. Measurements were based on ASTM E2638 (for closed rooms [4]) and ASTM E1130 (for open-plan offices [5]), with modifications to ensure consistency across office types. Sound sources were positioned at realistic talker locations, while receiver microphones were placed at typical seated ear heights, corresponding to respondent workstations or meeting room seating positions. Unoccupied background noise levels were measured at the same receiver locations with the sound source inactive. The same measured data were used to calculate SPC, PI, and SPP metrics.

Measurements covered multiple office type adjacencies, including open-to-open (both source and receiver located within the same open-plan office), closed-to-closed (source and receiver located in separate enclosed offices), open-to-closed, and closed-to-open configurations. This approach ensured that comparisons between metrics were based on identical acoustic conditions.

Key Findings

Figure 1 illustrates the relationship between subjective ratings and the three objective metrics for audibility, intelligibility, and distraction, with logistic regression curves fitted to the data. Across all office configurations and all three speech privacy descriptors, higher objective metric values for all three of the metrics investigated corresponded to improved subjective perceptions. However, clear differences emerged in how consistently each metric aligned with occupant responses across offices:

·         SPC showed the highest correlations with subjective ratings in both open-plan and closed offices, performing reliably across low and high privacy conditions.

·         PI and SPP showed greater variability relative to the logistic relationship curves, and slightly weaker alignment with subjective responses, particularly when applied across different office types.

Results demonstrate that SPC exhibits the strongest and most consistent relationship with occupant perceptions of the range of speech privacy descriptors across all office conditions. Table 1, therefore, presents the SPC scale derived from logistic regression analysis, along with the corresponding average occupant perception ratings for speech audibility, intelligibility, and distraction in non-secure speech privacy applications. It is emphasized that these classifications are based on averages and correspond to an estimated average vocal strength of 58.4 dBA (close to “normal” vocal strength [5]); appropriate acoustical design would provide sufficient safety margins when designing for the clear majority of office occupants [11].

Table 1. SPC scale with corresponding average occupant perception ratings for audibility, intelligibility, and distraction for non-secure speech privacy, based on a reference in-situ speech level of 58.4 dBA.

Figure 1. Relationship between subjective ratings and objective speech privacy metrics (PI, SPC, SPP) across all office configurations (open-to-open, closed-to-closed, open-to-closed, and closed-to-open) for audibility, intelligibility, and distraction.

Discussion

The findings demonstrate that SPC is the most robust metric for evaluating subjective speech privacy perception across both open and closed office environments. Unlike PI and SPP, SPC maintains sensitivity across a wide performance range and responds to changes in sound insulation and background noise levels, making it particularly suitable for design and diagnostic applications.

A key insight is the distinction between speech security and perceived speech privacy. SPC (per ASTM E2638 [4]) was developed primarily to assess secure speech conditions involving intentional eavesdropping. In contrast, this study captures the inadvertent experience of incidental speech intrusion during routine office work. As a result, occupants reported significantly better perceived privacy at given SPC values than the current interpretative categories in the standard suggest.

Conclusions

By directly correlating objective acoustic measurements with subjective occupant perceptions, this research demonstrates that SPC reliably predicts perception of audibility, intelligibility, and distraction in both open-plan and closed office environments.

Key conclusions include:

·         SPC outperforms PI and SPP in correlating with subjective speech privacy perception across different office configurations.

·         On average, occupants of typical offices experience substantially better perceived speech privacy at given SPC values than implied by the wording of the current secure speech-oriented ASTM E2638 standard [4].

·         The proposed unified SPC-based framework can effectively describe the average occupant perception of non-secure, everyday speech privacy in commercial offices.

Recommendations

Based on the findings, the following recommendations are proposed:

·         Adopt SPC as the primary metric for designing and evaluating speech privacy in both open and closed commercial office environments.

·         Revise interpretative guidance associated with ASTM E2638 [4]for non-secure, typical office applications to better reflect occupant perceptions.

·         Conduct further field validation in other office environments, such as healthcare or legal offices, where confidentiality requirements may heighten sensitivity to speech intrusion or lapses in speech privacy.

·         Integrate subjective evaluation into post-occupancy assessments to ensure that acoustic performance targets align with user expectations and experiences.

Acknowledgments

The research was supported by ASHRAE TC 2.6 through RP 1852 and an NSERC Discovery Grant, contributing to the development of performance metrics for speech privacy in high-performance buildings.

References

[1]     J. Kim and R. de Dear, “Workspace satisfaction: The privacy-communication trade-off in open-plan offices,” J. Environ. Psychol., vol. 36, pp. 18–26, 2013, doi: https://doi.org/10.1016/j.jenvp.2013.06.007.

[2]     M. Müller-Trapet, “Harmonizing Different Metrics for Speech Privacy,” in proceedings of the 23rd International Congress on Acoustics: integrating 4th EAA Euroregio, Berlin, Germany: Deutsche Gesellschaft für Akustik, Sep. 2019, pp. 2398–2405. doi: 10.18154/RWTH-CONV-239165.

[3]     A. Haapakangas, V. Hongisto, M. Eerola, and T. Kuusisto, “Distraction distance and perceived disturbance by noise—An analysis of 21 open-plan offices,” J. Acoust. Soc. Am., vol. 141, no. 1, pp. 127–136, Jan. 2017, doi: 10.1121/1.4973690.

[4]     American Society for Testing and Materials, “Test Method for Objective Measurement of the Speech Privacy Provided by a Closed Room (ASTM Standard No. E2638-24),” ASTM International, West Conshohocken, PA, Oct. 2024. doi: 10.1520/E2638-24.

[5]     American Society for Testing and Materials, “Test Method for Objective Measurement of Speech Privacy in Open Plan Spaces Using Articulation Index (ASTM Standard No. E1130-16 (2021)),” 2021. [Online]. Available: https://www.astm.org/e1130-16r21.html

[6]     W. Cavanaugh, W. Farrell, P. Hirtle, and B. Watters, “Speech Privacy in Buildings,” J. Acoust. Soc. Am., vol. 34, no. 4, pp. 475–492, Apr. 1962, doi: 10.1121/1.1918154.

[7]     E. Sundstrom, R. Herbert, and D. Brown, “Privacy and Communication in an Open-Plan Office: A Case Study,” Environ. Behav., vol. 14, no. 3, pp. 379–392, May 1982, doi: 10.1177/0013916582143007.

[8]     E. J. Ryerson and T. Rafferty, “Achieving speech privacy in challenging open office environments,” in InterNoise12, INTER-NOISE and NOISE-CON Congress and Conference Proceedings, New York City NY: Institute of Noise Control Engineering, 2012, pp. 7045–7054. Accessed: Oct. 23, 2025. [Online]. Available: https://ince.publisher.ingentaconnect.com/content/ince/incecp/2012/00002012/00000004/art00013;jsessionid=5659jbi86alku.x-ic-live-03

[9]     R. Bessey, “Planned versus achieved acoustical performance and occupant satisfaction for an office fit-out,” Proceedings of Meetings on Acoustics, vol. 19, no. 1, p. 040129, May 2013, doi: 10.1121/1.4800417.

[10]   A. Seddigh, E. Berntson, F. Jönsson, C. B. Danielson, and H. Westerlund, “The effect of noise absorption variation in open-plan offices: A field study with a cross-over design,” J. Environ. Psychol., vol. 44, pp. 34–44, 2015, doi: https://doi.org/10.1016/j.jenvp.2015.08.004.

[11]   R. Mackenzie, R. Toubar, and J. Lee, “Develop Performance Metric, Criteria, and Process to Measure and Predict Speech Privacy in High-Performance Buildings (Report No. 1852-RP). [Report in preparation],” 2025.