Cue tones need a spectrum offset from ambient noise

What it is

For a cue tone to be heard in a real environment, its own loudness is not enough — what matters is whether its frequency content overlaps with the frequency content of the ambient noise. If the tone's energy happens to sit right where ambient noise is already strongest, that noise will substantially raise its detection threshold; if the tone's energy instead sits in a band where the ambient noise is relatively weak, the same playback level leaves a much larger audibility margin. This turns the basic relationship — that frequency proximity drives masking — around into a design principle: actively avoid the frequency bands where noise concentrates, rather than simply cranking up the volume after the noise has already buried the tone.

Why it happens

Most common ambient noise (traffic, machinery, a noisy crowd) does not spread its energy evenly across the whole audible range; it tends to concentrate in the low-to-mid frequencies and thin out in the mid-to-high range. Deliberately placing a cue tone's energy in the band where noise is weak means the two activate overlapping cochlear regions much less, so the threshold elevation caused by the noise is far smaller — fundamentally the same frequency-proximity-drives-masking relationship, just exploited here for avoidance rather than passively suffered.

A single pure tone carries more risk: it has energy at only one frequency, so if that exact frequency happens to be covered by noise, the whole cue has no backup energy to fall back on. A complex tone with multiple harmonic components spread across several bands is more robust — even if one band gets suppressed by noise, the remaining bands may still cut through and be heard, effectively giving audibility a layer of redundancy.

Studying it

The usual approach is to first measure the ambient noise spectrum of the target usage scenario (via real recordings or spectral analysis with a sound level meter), then test a candidate cue tone's actual detection threshold or identification accuracy against that noise background under different spectral designs — comparing placing the tone's energy at the noise's peak band versus at its trough. This methodology is used systematically in developing industrial and medical alarm design standards, to produce quantitative recommendations for an alarm's spectral range and harmonic structure.

In interface sound design this method can verify whether a new cue tone leaves an adequate audibility margin across candidate deployment scenarios (a quiet room, a noisy street, inside a car), rather than being judged only by listening at the designer's own quiet desk.

Where it stops holding

• This strategy assumes the target environment's noise spectrum is roughly predictable and reasonably stable. If the actual noise composition varies too much and offers no consistent weak band to exploit, spectral offsetting has no reliable gap to use, and the design has to fall back on raising level alone or adding other redundant channels.
• Broadband noise (energy spread roughly evenly across the whole frequency range, such as certain wind noise or white-noise-like environments) has no clear "weak band," and the benefit of this strategy drops sharply.
• Spectral offsetting solves the problem of noise masking specifically — it cannot substitute for compensating a particular hearing-impaired population. If a target user already has severe hearing loss in some band, keeping the cue tone out of that band does nothing for that user's audibility elsewhere; that is a separate case requiring its own consideration.

Applying it

• Before deploying a cue tone, gather or estimate the target environment's noise spectrum, identify the band where noise energy is relatively weak, and place the tone's fundamental and main harmonics there — rather than picking whatever frequency happens to "sound crisp."
• Prefer a complex tone with a rich harmonic structure over a single pure tone; even if one harmonic falls into a noisier band, the remaining harmonics still provide backup audibility.
• Since common machinery and traffic noise concentrate in the low-to-mid range, placing a critical cue's main energy in the mid-to-high band (while avoiding the extreme high end where the ear's own sensitivity is already weak) typically cuts through this kind of ambient noise better than a low, deep tone.
• How to check: overlay the candidate cue tone on recorded or live ambient noise from the real environment, and have participants report the playback level at which the tone just becomes audible or identifiable, comparing it to the baseline level measured in quiet — to confirm the margin the spectral design actually delivers, rather than relying on the designer's own subjective impression.

Related

• Same group: A3.04.1 A loud sound masks a weaker sound close to it in frequency · A3.04.2 Masking extends forward and backward in time
• Nearby: A3.07 Alarm distinguishability (the full alarm-system design of audibility, differentiation and priority — this entry covers only the specific tactic of spectral offset from ambient noise) · A3.09 Ambient noise and signal-to-noise ratio
• Search terms: alarm design · spectral masking avoidance · Patterson alarm · warning sound audibility