Augmented control of hands free voice prostheses

Laryngectomy patients often use an electrolarynx to facilitate speech following a tracheotomy. Devices of this type provide the most intelligible means of communication for tracheotomy patients. However, the electro-larynx has inherent drawbacks such as the buzzing monotonic sound emitted, the need for a free hand to operate the device, and the difficulty experienced by many tracheotomy patients in adapting to use it. The most effective means of addressing the shortcomings of existing electro-larynges is to provide the user with a hands-free facility. This allows the user to perform other manual tasks whilst speaking, or simply to communicate more effectively through body language. Hands-free devices do exist but require a considerable amount of patient training as they involve the use of the shoulder muscles to control pitch. Furthermore, they are not suitable for all patients as the hands-free is suitable only for users with a certain type of tracheotomy. Goldstein et al in 2004 [1] produced a working prototype of a hands-free device that employed electromyographic signals to activate the device. However, it was quite cumbersome in design and failed to alleviate the monotonous sound produced. The goal of this research is to research the implementation of a hands-free electrolarynx, using various activation methods including electromyographic signals to vary parameters of the output signal. Once a satisfactory system of initiation has been devised and tested, a method of pitch variation shall be developed.