Take the 2-minute tour ×
User Experience Stack Exchange is a question and answer site for user experience researchers and experts. It's 100% free, no registration required.

Is there a standard for the frequencies and or duration used for error or alert beeps?

When I hear a ding or bing sound, I sometimes find that the frequency of the tone is too high or low (in my opinion) for the intent. For example, a high pitch tone a second long indicates an error to me, whereas a lower tone and quicker bing sound indicates a message.

Another example is that my car makes the same repeated bing sound when I turn the ignition to remind me about an engine service as it does when the outside temperature is close to freezing or I've not fastened my seatbelt.

This is clearly subjective on my part. Are there any standards in use or defined, either within a limited field, or more widely adopted for the frequency and or duration of tones used to indicate errors, warnings, and informational messages?

share|improve this question
1  
Not quite an answer - rather more questions: boxesandarrows.com/view/… –  katDNA Nov 21 '11 at 11:34
    
Excellent question. There's some existing research on this kind of thing in relation to industrial settings: nrc-cnrc.gc.ca/eng/ibp/irc/ci/volume-6-n4-1.html –  PhillipW Nov 21 '11 at 11:55
    
This has now got me wondering the history of how apple came to make their computers go 'bong' when they started up... –  PhillipW Nov 21 '11 at 12:36
add comment

2 Answers

up vote 2 down vote accepted

US Department of Defense Design Criteria Standard: Human Engineering, better known as MIL-STD 1472, has a pretty complete section (5.3) on audio signals. A quick read yields the following standards that seem to bear directly on your question (edited here):

5.3.1.3 Signal meaning. Each audio signal shall have only one meaning.

5.3.1.4 Apparent urgency. The attention gaining characteristics of the signals in a set (e.g., rapidity of pulse pattern, frequency, intensity) should match the relative priority of the signal.

5.3.3.1.1 Range. The frequency range shall be between 200 and 5,000 Hz and, if possible, between 500 and 3,000 Hz.

5.3.4.2.1 Attention and avoidance of startle reaction. To minimize startle reactions, the increase in sound level during any 0.5 sec period should be not greater than 30 dB. In addition, the first 0.2 sec of a signal should not be presented at maximum intensity, use square topped waveforms, or present abruptly rising waveforms

5.3.4.3.1 Use of different characteristics. When several different audio signals are to be used to alert an operator to different types of conditions, discriminable differences in intensity, pitch, beats and harmonics, or temporal patterns shall be provided. If absolute discrimination is required, the number of signals to be identified shall not exceed four. Signal intensity shall not be used alone as a means of discriminating between signals.

It discusses in detail characteristics of warning signals but keep in mind the military defines “warning” as, “you need to respond now or you’re going to die.” Not everything will apply to your case, I don't think.

It also provides guidance on whether you should have audio signals at all and what controls to provide for users.

The latest Windows UX Interaction Guidelines also provides guidelines on the use of sound (p686). They emphasize that all sounds provide only supplemental information -that users should never need to rely on sound alone for feedback. Thus, all sounds, including those for errors and warnings, should be brief, subtle, and calm, if they're present at all.

Use sound only when there is a clear advantage. When in doubt, don’t use sound... The ideal sound effect is one that users barely notice, but they would miss if it were absent.

As for the quality of each sound, that's more art than science. The Windows guidelines recommend:

Light, pure tones, and glassy and airy sounds, with a soft fade-in and fade-out (soft “edges”) to prevent abrupt, jarring, percussive effects. They are designed to feel subtle, gentle, and consonant. Windows sounds use echo, reverb, and equalization to attain a natural, ambient feel.

Simple beeps and buzzes of yesterday's computers are right out. You should exploit all the advanced sound reproduction capabilities found in modern systems. This means you need to create your sounds like you're creating a one or two note piece of music. If fact, maybe you should hire musician skilled with digital synthesizers, describe the information and effect you want to convey, and let him/her create some candidate sounds for you.

share|improve this answer
add comment

Higher pitched sounds about the 1KHz range are more unpleasant and therefore draw more attention and are therefore used more for alerts that need immediate attention. Think of a fire alarm here.

Lower pitched toned about the 100Hz range are more pleasant and therefore used more for notification tones. Computer startup sounds are examples.

Then, the faster the beep, the more annoying and hence alerting. It rarely makes sense using a beeping sound for a notification, so they are usually used more for attention grabbers.

finally if you really need to draw attention, you can use two discordant or dissonant tones together. This is usually only used for things like ambulances or emergency sirens.

Summary: make the tones more annoying to draw attention, and less annoying when it is simply a notification.

I can't recall the exact source for this but it was part of an audio engineering course at university.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.