Mozhi

Communication aid using approximate syllables instead of accurate words

The software has been developed specifically keeping in mind the limitations faced by people with spasticity. The philosophy behind the device and its functioning is that during verbal communication both the ‘speaker’ as well as the ‘listener’ participate in the communication process. It is this participation on the part of both parties that enables people with different accents and people with lesser command over the language, to communicate with others. Therefore the device works on the principle that it is sufficient to say words approximately – that there is no need for accuracy in pronunciation as a prerequisite for verbal communication.

The software is based on the identification of a limited set of sounds, using most words can be stated approximately. Using a list of commonly used word in English, it was found that most words can be stated to a fair degree of approximation using just a few sounds from a list of 12 to 16 sounds.

Since the software is based on the sound produced, and is not linked to any specific alphabet, the device can be used to communicate in any language. In addition, these sounds can be recorded using any person’s voice and therefore the user can speak with that persons voice – more specifically, if the user can produce even uncoordinated sound, the sound can be recorded and the specific segments of the sound isolated that correspond to the selected set of 12 to 16 sounds. Using these sounds, recorded from the user’s voice, it will be possible to enable a person affected with spasticity to speak in their own voice.

In addition to the sounds, a shortlist of most commonly used words have also been identified, and these can be used by the users as necessary, to enable faster communication.