The new system was created by researchers at Loughborough University in England. It is called the Augmentative and Alternate Communication device (AAC) and consists of a mask that is worn by the patient and linked via a computer to a speech synthesizer.
The team has seen a 97.5 percent success rate during early testing. The difficulty lies in training the computer to accurately recognise the “breath patterns” and then form them into the associated words.
The patient is able to specify what patterns to use for each word. The software can then detect and predict the signals that are likely to represent other phrases, learning from experience as it gains more usage.
AAC could represent the next step in giving paralyzed people a voice. Although some people have found success using computers that can accurately detect the slightest movement in a muscle or tiniest flicker of an eye, the condition of some patients means that even these techniques are not an option.
Dr. Atul Gaur, a consultant anaesthetist who worked with the team behind AAC, noted that breathing is an “almost effortless act which requires no speech, limb or facial movements.” He said: “This device could transform the way people with severe muscular weakness or other speech disorders communicate.”
The device is also much faster than other previous attempts at AAC products. It continually collects the analogue signals from the breath patterns and passes them through an analogue-to-digital converter for the computer.
Dr. David Kerr, Senior Lecturer at the School of Mechanical and Manufacturing Engineering, said this usage of continual analogue monitoring “should give us a greater speed advantage because more information can be collected in a shorter space of time.”
The invention is not the first of its kind though. Last year, 16-year-old Arsh Shah Dilbagi was placed as one of 15 finalists at the 2014 Google Science Fair after creating an AAC system that also uses human breath as its input.
Called Talk, the device can be produced for around $80 and is based on a micro-electrical-mechanical system (MEMS) – essentially an advanced microphone made of a diaphragm etched directly onto a silicon microchip.
Unlike Loughborough University’s device, the user of Talk does not need to wear a mask, instead placing a cord around the face. It interprets breaths as Morse Code by listening for two distinguishable exhales. The Morse Code can then be converted into words and sentences.
Once the breath patterns have been recognised and converted to words, both systems relay them to a speech synthesizer which can say the relevant sentences aloud. Systems like these give those who previously had no voice a chance to talk even when they have no control over the required muscles, an example of how empowering technology can be when adapted to suit people with disabilities.
