Scientists are developing a cochlear implant which could allow deaf people to hear music.
Scientists are trying to develop a more sophisticated implant
Existing implants mean people listen more easily to speech than music.
But a team at the UK's National Physical Laboratory have developed a device with a wider frequency range, which improves musical appreciation.
New Scientist magazine reports the whole implant could be put into the ear - current models require people to wear a box behind their ears.
The cochlea in the ear contains fluid and hairs which vibrate in response to sounds.
These hairs can stop vibrating, meaning people go deaf.
It can happen at any age but can be particularly difficult for children as it affects their ability to do well at school and socialise.
Cochlear implants currently involve putting an electrode inside the ear and an external box, which contains a microphone to pick up the sound, converts it from radiowaves into electrical signals and contains batteries to power the implant
Conventional hearing aids simply amplify sound rather than making it clearer.
The new implant resembles a comb, with a number of bar-shaped elements that vibrate in response to sound.
Each is tuned to resonate like a tuning fork at a different, narrow range of frequencies and is coated with a piezoelectric material - which creates an electrically generated pulse - so it does not require an external power supply.
By adjusting the length and diameter of the elements, the researchers have tuned each to resonate at a different, narrow range of frequencies.
When a sound, such as a musical note, causes one of the elements to vibrate, the flexing of the piezoelectric material produces a small voltage.
This is transmitted directly to the auditory nerve in the cochlea.
The prototype is still quite large, measuring two centimetres square.
They are working with researchers in the nanotechnology group at Cranfield University in Bedfordshire, UK, to develop a version which is significantly smaller, so it can fit into the cochlear.
Bill Nimmo, a member of the NPL team, said: "We would need 10 resonating elements for speech and 20 or maybe more for music.
"The challenge is to miniaturise the elements so that they still resonate at audible frequencies."
He said this means a commercial implant is likely to be at least 10 years away.
But once complete, the implant should enable people to have manual control over the frequencies they hear, enabling them to tune in to individuals in a crowded room and filter out the background chatter.
Angela King, of the Royal National Institute for the Deaf's audiology team, said: "Some people are too severely deaf to benefit from conventional hearing aids; a cochlear implant may be a suitable alternative but only for a proportion of them."
She said the device being developed by the National Physical Laboratory could be an advance - if it could work successfully at a much smaller size than the prototype.
"It shows huge promise, although we will not know for some years the true potential of the approach being investigated, it is in our view worth exploring to the full."