Principles of automatic processing of speech signals and their application in medical technology and for aids for...

Digital processing of speech signals has made great advances in the last few years. This has brought about many new possibilities for man-machine communication in information technology. Automatic recognition and synthesis of speech signals give chances to communicate in a very direct manner with data processing machines. This gives quite good advances in all areas where communication is often done by vocal commands. Medical technology is one example where people normally use spoken commands to get any action or to produce medical documents by voice input. Another and today still much more important area are aids for handicapped people. There are different tasks which may be done using speech recognizers or synthesizers. For paraplegics it will be important to operate their usual aids through vocal commands. For people with communication defects speech recognizers may be useful for the deaf to aid them in participating in speech communication or for training their speech capabilities. For people with speech defects through e.g. spastic paralysis speech aids may help them to produce an intelligible speech signal from written texts, produced by typing on a keyboard. Blind people may use reading machines with speech output to become relatively independent in text reading and text preparation using typewriters. Speech understanding systems may help to make a direct voice dialogue with medical expert systems. On the other hand, advanced research in hearing psychology and the physiology of speech production may help to develop more advanced automatic speech processing.