5.1 The speech privacy between locations in an open plan space is determined by the degree to which intruding speech sounds exceed the ambient sound pressure levels at the listener's ear; a classic signal-to-noise ratio situation.

5.2 The sound pressure levels at the listener's ear from intruding speech depend upon:

5.2.1 The individual vocal effort and orientation of the talker,

5.2.2 The attenuation of speech signals due to distance or intervening barriers, and

5.2.3 The reinforcement of speech signals due to reflections from surfaces such as the ceiling, furniture panels, light fixtures, walls, or windows.

5.3 The ambient sound levels within a space often must be increased in order to mask intruding speech using an electronic sound masking system. However, in certain locations and in specific frequency ranges, the building mechanical, electrical and plumbing (MEP) equipment, and the heating, ventilating, or air conditioning equipment (HVAC) may increase ambient sound levels or add tonal noise components that may require mitigation before tuning the masking sound.

5.4 The primary purpose of this test method is to assess the speech privacy for an average speech spectrum using the standard Articulation Index method. This requires measurement of the relevant acoustical characteristics discussed in 5.2 and 5.3 for a pair of locations and calculation of the Articulation Index using an average speech spectrum. The average speech spectrum is for male talkers speaking with normal voice effort. In specific cases such as designated quiet work zones for 'focused work' where administrative measures have been taken to reduce speech levels, a 'casual' voice spectrum should be used to calculate speech privacy, whereas in designated group work zones for 'collaborative work' where lively discussion is expected, a 'raised' voice spectrum should be used to calculate speech privacy.

5.5 The Articulation Index ranges from a low value of 0.00, where speech is generally perceived to be unintelligible, to a high value of 1.00, where all individual spoken words can be understood. Caution should be exercised in interpreting the numerical AI results of this test method since the percentage of single words, phrases, and sentences understood will be different for the same AI value (5).

5.6 This test method can be used to:

5.6.1 Compare the relative speech privacy afforded between different locations within open plan spaces.

5.6.2 Evaluate how changes in open plan components (barriers, furniture, ceilings, masking sound, or wall panels) affect speech privacy.

5.6.3 Assess speech privacy objectively for correlation with subjective responses.

5.7 This test method could be one element of a performance or acceptance test procedure. However, many additional items would need to be specified to allow the use of this test method for performance testing of an open plan environment, such as, the number of locations to be tested and method of selecting those locations, and the method of assessing the results. Specifying a numerical criterion in terms of the Articulation Index is also necessary for acceptance testing; however, the selection of such a criterion and permissible deviations should be undertaken with care in view of the present state-of-the-art as discussed in 5.5.