Optimal alarm sound design


The goal of this research is improving alarm sound design, focusing on the paradox of noticeable versus pleasant sounds. The characteristics of annoying sounds correspond in large extend to the characteristics of noticeable sounds. Therefore it is difficult to design an alarm sound, which is noticeable, but not annoying. A lot of studies are conducted along aspects which make sounds annoying and several guidelines are described for designing noticeable alarm sounds.

However, no data is conveyed about the combination of these characteristics linked to sound design. In this study an existing design process for alarm sounds is adjusted regarding to pleasantness and noticeability of sounds. Hereby methods to analyze and test pleasantness and noticeability of sounds are added to the process. As a result this paper suggests a new design process which can be used to design an alarm sound considering these aspects. Keywords product sound design; alarm sound; annoying sounds; noticeable sounds; design processes


Most of the time alarm sounds are not optimal designed, because in the majority of cases the context of the user is not taken into account. Edworthy (2006) found that in consequence alarms are often too badly designed resulting in annoying, not effective sounds. Schmidt & Baysinger (1986) pointed out that a pleasant sound to report a complication can be more effective during an emergency. However, the alarm sound should still be clearly noticeable, so that it cannot be missed. Annoying sounds The perception of annoyance may be very personal and subjective, but research as shown that there are some characteristics of sounds that influence this perceived annoyance. (Steele & Chon, 2007). The research of Steele & Chon (2007) found that loudness is the most important determinant of annoyance inrespectto sound. They also revealed that the wider the bandwidth, the more annoying the sound is perceived. Higher frequencies and modulations increase the perception of annoyance as well (Genuit, 2001). A modulation is a change from one tone to another. Besides the type of sound, there are a many other factors that influence the perception of annoyance. One of these factors, revealed by Maris et al. 2007), is the ability to influence the sound. Another study posited that age is also an important aspect for determining the perception of annoyance. (Botteldooren & Verkeyn, 2002). Noticeable sounds Obviously, the noticeability of an alarm sound is better when the volume of the sound is higher. Edworthy (2006) pointed out some other characteristics of clearly noticeable alarms sounds, like high and low frequencies, harmonics and discontinuous sounds. Hereby alarm sounds are respectively easier to localize, more resistant to masking by other sounds and less presumable to interfere withcommunication.

Harmonics are sounds with frequencies that are a multiple of the fundamental frequency. Another component which makes a sound more clearly noticeable, is the variation in more than only pitch (Edworthy & Meredith, 1997). Namely, if a sound alters in more than just tone, for instance frequency, the ability to distinguish it from other sounds increases. Besides that, environmental sounds and auditory icons are easier to learn and retain (Leung, 1997; Ulfvengren, 2003). Similarities annoying and noticeable sounds

There is a lot of literature written about designing noticeable alarm sounds and about annoying sounds. However, there is barely literature which compares these two characteristics of sounds. Nevertheless, a lot of characteristics of annoying sounds are identical to the characteristics of noticeable sounds, see figure 1. Figure 1. Characteristics which make sounds annoying as well as noticeable. As you can see in this figure, loudness is an important characteristic of annoyance as well as noticeability of sound. The same is true for high and low frequencies which are linked to a wide bandwidth and a high frequency.

High frequency is also linked with harmonics, because a harmonic is a sound whose frequency is higher than the keynote of that sound observed by the ear. Furthermore, modulations contribute to an annoying sound, while discontinuous sounds provide a noticeable sound. This is conflicting, because if a sound has modulations it is not discontinuous as a result of the changes in tone. So most of the characteristics of annoying and noticeable sounds are similar, whereby it is hard to design an alarm sound that is noticeable, but not annoying. All the corresponding characteristics are focused on type of sound.

However, there are other factors which can influence the perception of annoyance and the noticeability of sound. These factors are particularly related to the context. Research of Philip (2009) also showed the relationship between the annoyance and noticeability of sound. She showed a significant correlation between the urgency rating and the annoyance rating of alarm sounds. The urgency rating can be linked to the noticeability of the alarm sound, which in essence points towards a relationship between the annoyance and noticeability of a sound. Yet, there are no papers with guidelines or methods to deal with this problem.

Purpose The overall goal of this research is improving alarm sound design, focusing on the balance between noticeable and pleasant sounds. The aim of this research is suggesting a design method/process which helps to create a balance between the noticeability and annoyance of sounds. In this paper will be focused on a wide range of alarm sounds. Every sound with a warning function is taken into account. Hereby the outcome is usable for a variety of alarm sound designs, see figure 2. The IC is a relevant example, because noticeability is very important in this place, but if an alarm is too annoying it is often turned off.

Another example is an alarm clock. It is imported that the product helps you to wake up in time, but if the sound is not pleasant you may have a bad start to the day. Figure 2. Examples of products with alarm sound. For this research a literature study is done about existing methods which can be used to design alarm sounds. With these inputs and information about the annoyance and noticeability of sounds a new method is defined.


Design processes are most of the time iterative and consisting of different stages (Roozenburg & Eekels, 1995).

Often, a design process starts with a problem definition, followed by ideas to solve this problem. Then a concept is developed to meet the determined requirements and satisfy the user. (Rouse, 1991). Design process alarm sounds Edworthy and Stanton (1995) came up with a user-centered method to design alarm sounds, see figure 3. In this method, especially the noticeability of the sound is taken into account. In the method nothing is stated about the annoyance of sounds, though. From this process some stages which are also usable to design a pleasant, noticeable alarm can be derived.

Important stages regarding these aspects are establish the need for warning, appropriate ranking test, learning and confusion test, urgency mapping test, recognition and matching test and operation test. Figure 3. Edworthy and Stanton’s design process to design alarm sounds. The mentioned stages are particularly useful to design a noticeable sound. Therefore some stages that focus on the pleasantness of sound can be added. Besides that, the context of the users is not taken into account. However, this is very relevant to design an alarm sound, since most characteristics which are not conflicting are related to the context.

Context Another method which can be used to design alarm sounds is etnography. Ethnography is a method to define the context by revealing the user’senvironmentand interactions between the product, user and their own physical environment (Leonard & Rayport, 1997). This is useful for alarm sound design, since it is important to take the environment of the user, including other sounds, into account. Pleasantness There are some known techniques for eliciting verbal attributes of product sounds, which can be used to identify the importance of pleasantness of a sound.

One of these methods is the Repertory Grid Technique, succesfully applied by Berg & Rumsey (1999) for eliciting descriptions of the sound of a product. Another procedure to evoke important attributes of sounds is the Quantitative Descriptive Analysis (Stone et al. , 1974), whereby a descriptive language will be developed by participants accompanied by a facilitator. While applying these techniques no real sound examples are used, but products are presented to recall the attributes of sound connected to that product. Later in the design process eliciting can be used to rank the different sound designs by annoyance.

This can be done by deriving verbal attributes through a survey. Gabrielsson (1979) used this method by conducting an extensive experiment, where the participants were asked to rate the sounds according to their suitability. Another method to test the pleasantness of a sound is a model made by Aures (1985). This model calculates the pleasantness considering sharpness, roughness, tonalness and loudness, with an accuracy of more than 90%.


Based on the design process of Edworthy & Stanton (1995) a new design process can be suggested for designing pleasant, but noticeable alarm sounds, see figure 4.

Establish need for warning As the process of Edworthy & Stanton, the new design process will start with establishing the need for warning by identifying the alert functions of the product. This will be done to clarify the importance of noticeability of the sound, which can be used to create an optimal balance between noticeability and annoyance. For instance, when noticeability is very essential, the annoyance of a sound may be less important. Identify context Then, the context will be identified by using ethnography. Hereby the physical environment of the user and the corresponding sounds are taken into account.

Factors of the context can influence the annoyance and pleasantness of a sound. Therefore it is important to be aware of the context and use this during designing. Elicitation The last step of the analyis is elicitation of sounds, whereof users think they fit the product. By means of this method the importance of a pleasant sound can be revealed. There are two procedures which may be appropriate to do so. The Repertory Grid Technique can be used to elicit descriptions of the sound of a product. The more the word pleasant is called as description, the more important this characteristic will be.

Besides that, the Quantitative Descriptive Analysis is useful to develop a descriptive language which describes the desired attributes of a product. The outcome of this technique can be compared with those of The Reportory Grid Technique. The difference is in the way the descriptions are elicit. The Reportory Grid Technique uses product sounds and the Quantitative Descriptive Analysis uses no real sounds, but only products. Figure 4. Suggested design process regarding pleasant, but noticeable alarm sounds Designing sounds The design of the sound can be done exactly the same as Edworthy and Stanton escribed in their process. First, existing alerting sounds and user suggestions will be considered. After that soundimagery studies can be used. Testing pleasantness/annoyance After designing different concept sounds, the sounds can be tested in terms of annoyance or pleasantness. To do so two methods are suggested. The first procedure consists of a questionairre or survey, whereby the participants should rank the sounds by annoyance. The other technique is a model developed by Aures (1985) which calculates the sensory euphony(pleasantness) of a sound regarding loudness, sharpness, tonalness and roughness.

According to these tests the sound can be modified considering pleasantness. Testing noticeability The noticeability of sounds can be tested by procedures Edworthy & Stanton (1995) described in their design process. Using these techniques the sounds can be tested concerning different aspects, like learning & confusion, urgency and recogniton & matching. The learning & confusion test is ideally performed in the real context, so it will become clear if the sound can be confused with other sounds in the environment.

The urgency mapping test shows if the mapping between the signal and situation is suitable in terms of urgency using existent guidelines. In the recognition & matching test participants are asked to assign the sounds to the appropriate warning function. According to these tests the sound can be improved and optimized considering noticeability.


The adjustments of the suggested design process are really focused on the analysis and testing of pleasantness and noticeability. Consequently, the designing itself is not taken into account, but is indeed very important.

Further research for this phase of the design process can be useful to improve the suggested design process. Besides that, the design process is not tested, whereby there is not been evaluated how useful the process might be. To validate the suggested process an experiment with a couple sound designers should be done. Furthermore two procedures who explore roughly the same are recommended for the elicitation phase, but only one technique is necessary in this phase. Therefore more research about these methods in regarding to pleasantness and noticeability is preferred to choose the most appropriate technique.


Due to the fact that most characteristics of noticeable sounds are conflicting with the characteristics of pleasant sounds, it is difficult to design a sound which is both noticeable and pleasant. However, if the designer will focus on this during the whole design process, it should be possible to find a good balance between these two aspects. Hereby it is essential to look at the context of the user and the need for warning and a pleasant sound. Furthermore the sound should be tested on these aspects, since the characteristics are subjective and therefore defined by the user.

The suggested design process can be used for a wide range of alarm sounds, which requires a noticeable as well as a pleasant sound. Some examples of such alarms are alarm clocks, microwaves and alarms in intensive care units.


  1. Aures, W. (1985). Berechnungsverfahren fur den sensorischen Wohlklang beliebiger Schallsignale. Acustica, 59: 130-141
  2. Berg J. and Rumsey, F. (1999). Spatial Attribute Identification and Scaling by Repertory Grid Technique and other Methods. Proc. AES 16th International Conference
  3. Botteldooren, D. and Verkeyn, A. (2002). Fuzzy models for Accumulation of reported community noise annoyance from combined sources, Journal of Acoustic Society of America, 112(4): 1496 – 1508
  4. Edworthy, J. and Stanton, N. (1995) A user-centered approach to the design and evaluation of auditory warning signals: 1, Methodology, Ergonomics, 38(11): 2262-2280.
  5. Edworthy J. and Meredith C. (1997). Influence of verbal labelling and acoustic quality on the learning and retention of medical alarms. Int J Cogn Ergon; 1: 229–43
  6. Edworthy, J. and Hellier, E. (2006). Alarms and human behaviour: implications for medical alarms. British Journal of Anaesthesia, 97(1): 12–17
  7. Gabrielsson, A. (1979). Dimension analyses of perceived sound quality of sound-reproducing systems. Scand. J. Psychol. 20: 159-169
  8. Genuit, K. (2001). The problem of predicting noise annoyance as a function of distance. Internoise, Rome, Italy.
  9. Leung YK, Smith S, Parker S, Martin R. (1997). Learning and retention of auditory warnings. In: Frysinger S, Kramer G, eds. Proceedings of the Third International Conference on Auditory Display (ICAD), Palo Alto, CA, USA. Available from http://www. santafe. edu/_icad__
  10. Maris E. , Pieter J. , Stallen, P. J. , Vermunt R. , Steensma H. (2007). Noise within the social context: Annoyance reduction through fair procedures, Journal of Acoustic Society of America, 121(4): 2000 – 2010
  11. Philip, E. (2009). Evaluation of medical alarm sounds. Doctoral thesis, New Jersey Institute ofTechnology.
  12. Roozenburg, N. F. M. and Eekels, J. (1995). Product Design, fundamentals and methods. (Lemma BV, Den Haag. )
  13. Rouse, W. B. (1991). Design for success- A Human-Centered Approach To Designing Successful Products and Systems, John Wiley & Sons Inc. ISBN 0-471-52483-2.
  14. Schmidt S. I. , Baysinger C. L. (1986). Alarms: help or hindrance? Anesthesiology, 64: 654–5
  15. Steele, D. L. & Chon, S. H. (2007) A Perceptual Study of Sound Annoyance. Proceedings of the 2nd Audio Mostly Conference. pp. 19-24
  16. Stone, H. , Sidel, J. , Oliver, S. , Woolsey, A. , Singleton, R. C. (1974). Sensory evaluation by quantitative descriptive analysis. FoodTechnology, 24-34
  17. Ulfvengren P. (2003). Design of natural warning sounds in human-machine systems. Doctoral thesis, Stockholm Institute of Technology.