Noise Health Home 

[Download PDF]
Year : 2003  |  Volume : 5  |  Issue : 19  |  Page : 69--73

Observations of noise exposure through the use of headphones by radio announcers

W Williams1, J Presbury2,  
1 National Acoustic Laboratories, Chatswood, Australia
2 Australian Broadcasting Commission, Sydney, Australia

Correspondence Address:
W Williams
National Acoustic Laboratories, 126 Greville Street, Chatswood, NSW 2067


This study examined the potential risk of hearing loss by commercial radio announcers. This risk is developed through the regular use of headphones. These headphones are used to monitor broadcast transmission and communication information from program producers. To our knowledge there are no published studies of the noise exposure of radio announcers. The experimental method utilised a headphone parallel to the one in use mounted on a wideband, artificial ear. A Sound Level Meter was then used to measure the sound level and then calculate the exposure level. Depending on the feedback level applied to their headphones radio announcers are exposed to potentially damaging levels of noise. Levels measured correlate with results from other studies of long-term average speech spectrum and voice level measurements.

How to cite this article:
Williams W, Presbury J. Observations of noise exposure through the use of headphones by radio announcers.Noise Health 2003;5:69-73

How to cite this URL:
Williams W, Presbury J. Observations of noise exposure through the use of headphones by radio announcers. Noise Health [serial online] 2003 [cited 2020 Sep 21 ];5:69-73
Available from:

Full Text


It has long been suspected that individuals who work professionally in industries where they must wear headphones in order to listen to either incoming or outgoing communication signals are at risk of noise injury and a subsequent hearing loss.

Most users of communication headsets can be divided into two general categories. In the first category users, such as airline pilots and factory workers, are endeavouring to receive a communication signal in a noisy background and utilise noise excluding headphones to attenuate background noise and enhance the communication signal. The second category of users, such as call centre operators or "help" desk workers, are primarily trying to use a 'hands free' communication system where background noise is not normally a significant difficulty.

Radio announcers are a special case. They are specifically trying to monitor their own voice transmission for quality control purposes. They also simultaneously receive communication through the headphones from program producers. Headphone users such as those in the two categories mentioned above do not appear to monitor their speech loudness or quality in the same critical manner. Rather, experienced headphone users simply maintain a constant speech level through the normal use of the air­conduction path from the voice to the ear.

There are many studies reported in the literature on how to measure noise exposure under headphones (eg Macrae: 1995; Van Moorhem, Woo, Liu and Golias: 1996; and Dajani, Kunov and Seshagiri; 1996). Some work has been carried out on the noise exposure of airline pilots using headsets on noisy flight decks (Goodwin: 2001) and naval radio operators (Holmes: 1998). However, no references could be found on the potential noise exposure of radio announcers or indeed, on any noise exposure research involving radio announcers.


The method of measurement for noise exposure under the headphone was as per the requirements of Australian/New Zealand Standard AS/NZS 1269.1: 1998, Occupational noise management Part 1: measurement and assessment of noise immission and exposure, Appendix C "Recommended procedures for measurement of sound pressure levels from headphones or insert earphones".

This method involved the use of an identical headphone connected in parallel with the headphone in use. The parallel headphone was placed on a wide-band artificial ear, a B&K type 4152, that was connected to an integrating Sound Level Meter, B&K type 2231. The L' Aeq ,T measured under the headphone was then adjusted by -8 dB as described in Macrae (1995) and AS/NZS 1269.1: 1998. This -8 dB correction accounts for the transfer function from the coupler levels to produce the equivalent free field sound levels at the listener's ear and allows the level under the headphone to be compared to exposure risk criteria as normally expressed in occupational health and safety regulations.

The parameter measured was the A-weighted, equivalent, continuous sound pressure level over a sample time T, L' Aeq,T . This parameter can then be simply compared to the continuous, A­-weighted exposure level,L Aeq,8h .

Measurements were taken during broadcast times while the radio stations were conducting typical daily, live transmissions from their normal operating studios. The procedure did not interrupt normal activities and the announcers were able to continue their work without interruption or disturbance.

Sample times varied from a few minutes, typical for a news broadcast, to several hours representing an extended morning, afternoon or evening broadcast. Sample times were chosen to represent the program under study. For example a "news" broadcast may have been sampled for the whole ten minutes of "news" presentation while a morning show may have been sampled for several periods of fifteen minutes each, over the total broadcast time of up to three hours.

A cross-section of professional radio announcers in New South Wales was selected in an attempt to form a representative sample. Eight males and four females distributed over both city and regional stations were evaluated. Programs were typical of a daily selection that would be found across Australia and included talk-back, interviews (live, pre-recorded and telephone), news (local and syndicated), music (from classical to heavy rock) and commentary (live and pre-recorded).


The results are summarised in [Table 1]. Average noise levels vary widely across individual announcers. There is no clear effect of program content on noise levels.

Statistical analysis of the above results shows that there is no significant difference between male and female announcers (t=0.27, df=10, p=0.79). However, noise levels were significantly higher for the four metropolitan announcers (mean= 812 dB, s=6.0) than for the regional announcers (mean=68, s=11.0) with t=2.46, df=7 and p=0.04)


In work concerning the prescription of hearing aid gain Cornelisse, Gagne and Seewald (1991) found that the mean overall SPL of the long-term average speech spectrum (LTASS) at ear level (American-English speakers) was 70 dB for females and 71 dB for males. The levels measured at 30 cm in front of the mouth were 65.1 dB and 67.4 dB respectively. For a mixture of male and female subjects Byrne et al (1994) measured a LTASS value of 72 dB SPL at a distance of 20 cm.

From the results it can be seen that the long term average levels under the headphones are much higher that would normally be expected considering that the measured background noise in the typical studio varied between 30 dB and 35 dB SPL. This implies an average signal to noise ratio (S/N) of between 45 dB and 50 dB for participants in this survey.

It is "accepted that a S/N of 12 dB is usually adequate to reach a normal-hearing individual's threshold of intelligibility" (Berger, Royster, Royster, Driscoll and Layne: 2000: p 587) following work by Hawkins and Stevens (1950) on the masking of speech. Why then should radio announcers choose a S/N of 45 dB to 50 dB or greater? In fact the actual range for S/N is in the order of 30 dB to 65 dB.

One explanation is that the announcers are not trying to hear what is coming through the headphones over the background noise, but rather trying to hear the headphone signal over their own voice level.

Hain, Burnett, Larson and Kiran (2001) examined the effects of auditory feedback on speech. Participants were instructed to vocalise at a comfortable level throughout the experiment and to maintain a constant vocal loudness of around 70 dB SPL. Vocal level was maintained by the use of voice feedback through electronic headphones. To maintain this vocal loudness of 70 dB SPL, voice feedback loudness at the ear was maintained at approximately 10 dB above the vocal loudness, ie at around 80 dB (p 2147).

This 10 dB difference is consistent with the vocal loudness (L Aeq,T ) at the ear of announcers' voices measured in the current study. These were 78 dB for males, 80 dB for females and 79 dB overall. These are approximately 10 dB greater than figures obtained by Cornelisse, Gagne and Seewald (1991) of 71.3 and 70.2 for males and females respectively for normal conversational speech.

If individuals, such as radio announcers, use headphones to monitor their own voice feedback in order to maintain vocal quality and loudness why is a S/N of 10 dB not sufficient for all individuals?

Some announcers experienced noise levels that represent a potential risk, for example those with L Aeq,T measures of 92 dB, 95 dB and 91 dB. An L Aeq,T of 83 dB could be considered to be of marginal risk depending on the total exposure time while lower L Aeq,T exposures would be considered negligible. The degree of risk is a function of the L Aeq,T and the time period that the announcers use the headphones. The longer the time and the greater the L Aeq,T then the greater the risk. The minimum on air period for an announcer is usually two hours while some announcers can at times operate up to four or five hours. The exceptions to this are news readers who usually operate from only a few minutes up to a maximum of fifteen minutes at a time.

ISO 1999 - 1990 was established in order that the noise-induced permanent threshold shift could be determined for an adult population for various levels of noise and exposure times. While this International Standard does not and can not clearly define an exposure limit it is now generally recommended by most jurisdictions that the degree of "acceptable" risk is an L Aeq,8h of 85 dB (see Goelzer, Hansen and Sehrndt: 2001, p 81). The World Health Organization (1980) stated that "risk is negligible at noise exposure levels of less than 75 dB(A) L eq (8-h)" (p 14).


Based on this study it appears that while on average most radio announcers are experiencing equivalent at ear noise exposure levels that do not create risk, some announcers experience noise levels that represent a potential risk to their hearing health.

If radio announcers are using headphones as a normal part of their work practices then they should be aware of the potential risks to hearing.


The authors would like to thank Dr Suzanne Purdy and Dr Harvey Dillon for their constructive input to this work.[14]


1Australian/New Zealand Standard AS/NZS 1269.1: 1998 Occupational noise management Part 1: measurement and assessment of noise immission and exposure, Standards Australia, Sydney
2Berger, EH, Royster, LH, Royster, JD, Driscoll, DP and Layne, M editors (2000) The Noise Manual" AIHA Press, Fairfax, VA
3Byrne, D, Dillon, H, Tran, K, Arlinger, S, Wilbraham, K, et al (1994) An international comparison of long-term average speech spectra, J Acoustic. Soc Am, 96(4), 2108­-2120
4Cornelisse, LE, Gagne, JP and Seewald, RC (1991) Ear­Level Recordings of the Long-term Average Spectrum of Speech, Ears and Hearing,12(1), 47-54.
5Dajani, H, Kunov, H and Seshagiri, B (1996) Real-Time Method for the Measurement of Noise Exposure from Communication Headsets, Applied Acoustics 49(3), 209­-224
6Goelzer, B Hansen, CH and Sehrndt, GA, editors, (2001) Occupational Exposure to Noise: Evaluation, Prevention and Control A document published on the behalf of the World Health Organization, Federal Institute for Occupational Safety and Health, Dortmund
7Goodwin, T (2001) Flight Deck Noise, The Log a Publication of the British Airline pilots Association (undated) July 25, 2001
8Hain, TC, Burnett, TA, Larson, CR and Kiran, S (2001) Effects of delayed auditory feedback DAF) on the pitch­shift reflex, J. Accoust. Soc Am, 109, (5) Pt 1, 2146-2152
9Hawkins, JE and Stevens, SS (1950) The masking of Pure Tones and Speech by White Noise, J. Accoust. Soc Am, 22 (1), 6-13
10Holmes, J (1998) Noise exposure from communication headsets aboard HMS Illustrious, Institute of Naval Medicine Technical Report (Health & Hygiene) No 98028, Alverstoke, Gosport, UK
11 ISO 1999 - 1990 Acoustics - Determination of occupational noise exposure and estimation of noise­induced hearing impairment International Organisation for Standardisation, Geneva, 1990
12Macrae, JH (1995) Hearing Conservation Standards for Occupational Noise Exposure of Workers from Headphones or Insert Earphones, The Australian Journal of Audiology, 17(2), 107-114
13Van Moorhem, W, Woo, KS, Liu, S and Golias, E (1996) Development and Operation of a System to Monitor Occupational Noise Exposure Due to Wearing a Headset, Appl.Occup.Environ.Hyg 11(4), 261-265.
14World Health Organization (1980) Environmental Health Criteria 12: NOISE, Geneva