"Tonotopic organisation" is the way the ear sorts out the different frequencies so that our brain can process that information. High frequency sounds (which are high pitched sounds) do not pass very far along the cochlear, but low frequency sounds pass further in. In a normal ear, the sound causes vibration of hair cells which are located all along the cochlear. The movement of the hair cells creates an electrical disturbance that can be picked up by the surrounding nerve cells.
If hearing loss is caused by a problem in the hair cells or any other problem related to the changing of sound energy to nerve pulses, then the problem may be able to be rectified by a cochlear implant.
The implant is capable of creating an electrical pulse. It does this at a certain point along the cochlear. This electrical pulse is interpreted by the brain as a sound at a certain frequency.
|
The brain is addaptive. If someone has not had hearing for a long period of time the brain will addapt and use the area of their brain normally used for hearing for other functions. If a person has lost their hearing for a long time and then regains it using an implant, the sounds can be very disoriantating as the brain is reading these signals not as sonds but as other things.
The internal part of the device contains:
Waveform processing strategies use bandpass filters to divide the signal into different frequency bands. The algorithm chooses a number of the strongest outputs from the filters. The number depends on the algorithm, and can also depend on whether the sound is determined to be a consonant or a vowel sound.
Feature extraction strategies use features which are common to all vowels. Each vowel has a fundamental frequency (the lowest frequency peak) and formants (peaks with higher frequencies). The pattern of the fundamental and formant frequencies is specific for different vowel sounds. These algorithms try to recognise the vowel and then emphasise its features.
wikipedia.org dumped 2003-03-17 with terodump