linuxaudio.org

Simon Fielding wrote:

quoted text
> My sister is a specialist teacher for hearing-impaired primary school

> children. As part of her curriculum she includes music and in

> particular, nursery rhymes etc for the younger children. She would like

> them to be able to sing these at home with and/or for their parents. For

> those children with non-hearing-impaired parents, this is not a problem

> but many of the children also have hearing-impaired parents. Therefore

> she would like to produce a CD of her singing for the children to take

> home and use. I would really appreciate any help members can give me as

> I am not a professional musician or recording engineer.

>

> 1) She will be singing unaccompanied (she is a trained singer and is

> perfectly competent to do this) in an alto register to avoid any

> distraction for the children from accompaniments etc. Does anyone have

> any suggestions about this? (eg effects for recording, effects for the

> headphone mix etc)

For most hearing impaired listeners, less reverberation is usually

better (if understanding the words is desirable). The effect of

compression very much depends on the accompaniment and/or distracting

noises; in a clean situation compression usually makes things worse (see

e.g., R. Plomp: The negative effect of amplitude compression in

multichannel hearing aids in the light of the modulation-transfer

function. J. Acoust. Soc. Am., vol 83(6), 1988). Try to avoid pops

during recording.
>

quoted text
> 2) This one's a bit more specialist so you may not be able to help -

> hearing impairment often starts with loss of high frequency response.

> The obvious thing would seem to be to boost these but I don't know if

> that would be correct. Does anyone know?

If they will listen via loudspeaker and wear their hearing aids

(assuming that they have one), all individualized frequency shaping is

already done by the hearing aid. Listening via headphone is not really

working with hearing aids, thus an individual frequency shaping might be

beneficial. However, for that you need to know a bit about the

individual hearing (e.g., hearing threshold and whether it is

sensorineural or conductive loss, as a bare minimum). A very simple rule

of thumb is to apply a frequency dependent gain which is 40% of hearing

threshold for sensorineural loss and 100% of threshold for conductive loss.
Linux audio tools can be used to do batched pre-filtering of audio

material based on individual hearing loss (e.g., the command line tool

'applyplugin' which applies LADSPA-Plugins to files, can load equalizers

and set their gains).
Best regards,
Giso
>

quoted text
> Any comments on any other aspects of this project would also be more

> than welcome.

>

> Hopeful regards,

> Simon

> _______________________________________________

> Linux-audio-user mailing list

> Linux-audio-user@lists.linuxaudio.org

> http://lists.linuxaudio.org/listinfo/linux-audio-user

_______________________________________________

Linux-audio-user mailing list

Linux-audio-user@lists.linuxaudio.org

http://lists.linuxaudio.org/listinfo/linux-audio-user