Audio Output

For situations with poor viewing conditions e.g. low illumination or high vibration, audio output can provide another modality of information dissemination or provide more redundancy. Audio messages are most appropriate where an immediate response is required with less reliance on referal to the message at a later date. It is possible to use audio output to supplement environments where people are receiving high amounts of visual information even whilst a person is moving around. Audio information is useful when the information is continually changing, for warnings or signals of action. But it is not appropriate in environments with a large amount of background noise.

Loud tones undoubtedly attract attention but can also be startling, possibly leading to a stressful environment which will decrease information processing. The source of auditory information can also be difficult to differentiate causing confusion. The spatial dimension e.g. aimed at only one ear, or both ears can provide a means of differentiation. Pitch could increase the redundancy of these channels of information. The tone of audio information needs to be carefully considered due to the variability of background noise. The length of the signal should be between 100 and 150ms. The number of warning tones should be limited to below four distinguishable tones. These can be distinguished by their spectral content, pulse duration, pulse shape or temporal pattern. Varing the temporal pattern of successive tones helps them to be distinguished.Urgent warning sounds should be very distinct without startling people.

There are a large number of people with impaired hearing. Many of these people may have little problem in carrying on a conversation in a quiet room, but have significant problems using a public telephone in a noisy environment such as a railway station or airport.

It is estimated that there are over four million people in the European Union with a severe hearing impairment. Many of these people will want to use a hearing aid to give them full use of the telephone. Coupling to a telephone may be by any appropriate means, including inductive coupling, infra-red, direct electrical connection, or by matched acoustic transducers.

illustration of volume buttons on a handset Receiver amplification enables the user to increase the volume of sound coming through the telephone earpiece. This is helpful to people who either do not have a hearing aid or who need to remove their hearing aid to use the phone. This facility should allow for increasing the level of the acoustic output by user adjustment over a range of 0-20dB. For safety reasons it is recommended that the control should reset to zero on termination of a call.

A more convenient way of presenting display messages for those people who have difficulty in reading them is to make use of synthetic speech. A means of adjusting volume, and of initiating or repeating the message, will be required. Poor quality synthetic speech can be difficult to understand by those with a mild hearing impairment.

Image of ringer pitch and volume controls Ringer
People with a hearing loss that is significantly frequency dependent may need to be able to shift the dominant tone of the ringer into a region where their hearing acuity is higher. This is particularly important where high-pitched ringers are used, since the most common forms of hearing impairment affect the higher frequencies disproportionately. The maximum output level from the ringer should provide good audibility even in domestic environments where acoustic absorption may be high. A means of controlling this output will then be necessary. Only with the more severe forms of hearing loss should it be necessary to utilise separate high volume ringing devices or visual signallers.

Speech

Some guidelines for synthesized speech include:

Voice warnings should be presented in a voice that is different from other voices that will be heard in the task situation.
If synthesized speech is used for other types of information in addition to warnings, the user needs to be able to distinguish between these messages.
Maximise the intelligibility of the messages.
Make the voice as natural as possible so people are more likely to accept it.
If the message is missed, it is beneficial for people to be able to replay it.
If the message is familiar, the ability to interrupt the message would be benefiticial for experienced users.
Use synthesized speech where it is appropriate and acceptable to users.
Computer generated speech is appropriate for situations that require different messages to be generated.
Where the choice of messages is relatively limited, human voices are preferred because synthetic speech is less intelligible and less preferred.
Use nonspeech audio messages only for the purposes of alerting.
Make a headphone socket available for private speech output.
A volume control is beneficial.

Recommendations

ETSI standards (ETS 300 381 and ETS 300 679) should be observed where appropriate.
The sidetone level should be user adjustable within the limits set by the network characteristics.
Provide user control of volume of audio output.
For acoustic signals to attract attention, use a frequency between 300Hz and 3000Hz.
Messages should be simple and short.

Relevant standards

ETS 300 381 Telephony for hearing impaired people: Inductive coupling of telephone earphones to hearing aids.
ETS 300 388 (December 1994) Telephony for hearing impaired people: Inductive coupling of telephone earphones to hearing aids.
ETS 300 488 (January 1996) Telephony for hearing impaired people: Characteristics of telephone sets that provide additional receiving amplification for the benefit of the hearing impaired.
ETS 300 679 (September 1996) Telephony for the hearing impaired: Electrical coupling of telephone sets to hearing aids.
ITU P370 Magnetic field strength around the earcap of telephone handsets which provide for coupling to hearing aids.

Further information

Telephones - What features do disabled people need?
Wilson, J.R. & Rajan, J.A. (1999). Human Machine Interfaces for Systems Control. In J.R. Wilson & E.N. Corlett. (Eds.). Evaluation of Human Work. Taylor and Francis. UK.