Noise Masking Threshold Calculator
Whether you can hear a sound depends not just on its level but on the background noise around it. When the noise is loud enough relative to the signal, it masks it and the signal becomes inaudible, which is the principle behind office sound masking for speech privacy. This calculator compares a signal level against a noise level using an adjustable masking offset to report the signal-to-noise margin and whether the signal is likely audible or masked. Because the exact masking threshold depends on the situation, the offset is a user-editable input rather than a fixed figure.
Masking threshold formula
SNR = signal level - noise level
Masking threshold = noise level + offset
Margin = signal level - masking threshold
Audible if margin > 0, otherwise masked
The offset sets how far above the noise a signal must rise to be detected. Adjust it to match your detection criterion or measured just-audible data.
Masking context
- Positive SNR generally favors audibility; strongly negative SNR means the signal is masked.
- Speech intelligibility commonly wants +6 to +15 dB SNR.
- Office sound masking targets roughly 40 to 48 dB(A) of steady noise.
- Masking is strongest within the same critical frequency band.
- Low-frequency maskers can mask higher-frequency signals (upward spread).
Noise masking: frequently asked questions
What is auditory masking?
Masking is when a louder sound (the masker, often background noise) makes a quieter sound (the signal) inaudible or harder to hear. A signal is generally detectable when its level exceeds the masking threshold, which sits a small margin above the masker level in the relevant frequency band. Speech privacy systems and sound masking deliberately raise background noise to mask intruding speech.
How is the masking threshold estimated here?
This calculator uses a simple energetic model: the signal is considered audible when the signal-to-noise ratio (signal level minus noise level) exceeds a masking offset you set. The default offset of -3 dB reflects that a tone roughly at the noise level can still be detected; you can adjust the offset to match your criterion or measured data.
What signal-to-noise margin makes speech intelligible?
For reliable speech intelligibility, a positive signal-to-noise ratio is preferred, commonly +6 to +15 dB depending on the listening task. Near 0 dB speech is difficult, and a strongly negative margin renders it unintelligible, which is the goal of sound masking for privacy. Enter your levels to see the margin and the masking outcome.
Why use sound masking in offices?
Sound masking adds a low, steady, broadband background sound that raises the noise floor so that nearby speech falls below the masking threshold and becomes unintelligible, improving acoustic privacy. The added noise is typically tuned to a gentle spectrum and a level around 40 to 48 dB(A), enough to mask speech without being intrusive.
Does masking depend on frequency?
Yes. Masking is strongest when the masker and signal share the same critical band, and a low-frequency masker can also mask higher-frequency signals (upward spread of masking). This simplified calculator compares broadband or single-band levels; for detailed work, evaluate masking band by band.
Official sources
- NIOSH (CDC): Noise and Occupational Hearing Loss.
- NIST: National Institute of Standards and Technology.
Reviewed by the CalculatorHub team, edited by James Graham, 17 June 2026. See our methodology.