Automatic speech recognition and training for severely dysarthric users of assistive technology: the STARDUST project.
Parker M, Cunningham S, Enderby P, Hawley M, Green P. Sheffield Speech and Language Therapy Agency, Sheffield, UK. Clin Linguist Phon. 2006 Apr-May;20(2-3):149-56.
The STARDUST project developed robust computer speech recognizers for use by eight people with severe dysarthria and concomitant physical disability to access assistive technologies. Independent computer speech recognizers trained with normal speech are of limited functional use by those with severe dysarthria due to limited and inconsistent proximity to “normal” articulatory patterns. Severe dysarthric output may also be characterized by a small mass of distinguishable phonetic tokens making the acoustic differentiation of target words difficult. Speaker dependent computer speech recognition using Hidden Markov Models was achieved by the identification of robust phonetic elements within the individual speaker output patterns. A new system of speech training using computer generated visual and auditory feedback reduced the inconsistent production of key phonetic tokens over time.
Prescribing assistive-technology systems: focus on children with impaired communication.
Desch LW, Gaebler-Spira D; Council on Children With Disabilities. Collaborators (12) Murphy NA, Cartwright JD, Desch LW, Duby JC, Elias ER, Liptak GS, Myers SM, Norwood KW Jr, Sagerman PJ, Tilton AH, Lipkin PH, Gaebler-Spira D. Pediatrics. 2008 Jun;121(6):1271-80.
A speech-controlled environmental control system for people with severe dysarthria.
Hawley MS, Enderby P, Green P, Cunningham S, Brownsell S, Carmichael J, Parker M, Hatzis A, O’Neill P, Palmer R. Department of Medical Physics and Clinical Engineering, Barnsley Hospital NHS Foundation Trust, UK. firstname.lastname@example.org Med Eng Phys. 2007 Jun;29(5):586-93. Epub 2006 Oct 17.
Automatic speech recognition (ASR) can provide a rapid means of controlling electronic assistive technology. Off-the-shelf ASR systems function poorly for users with severe dysarthria because of the increased variability of their articulations. We have developed a limited vocabulary speaker dependent speech recognition application which has greater tolerance to variability of speech, coupled with a computerised training package which assists dysarthric speakers to improve the consistency of their vocalisations and provides more data for recogniser training. These applications, and their implementation as the interface for a speech-controlled environmental control system (ECS), are described. The results of field trials to evaluate the training program and the speech-controlled ECS are presented. The user-training phase increased the recognition rate from 88.5% to 95.4% (p<0.001). Recognition rates were good for people with even the most severe dysarthria in everyday usage in the home (mean word recognition rate 86.9%). Speech-controlled ECS were less accurate (mean task completion accuracy 78.6% versus 94.8%) but were faster to use than switch-scanning systems, even taking into account the need to repeat unsuccessful operations (mean task completion time 7.7s versus 16.9s, p<0.001). It is concluded that a speech-controlled ECS is a viable alternative to switch-scanning systems for some people with severe dysarthria and would lead, in many cases, to more efficient control of the home.
Speech recognition software as an assistive device: a pilot study of user satisfaction and psychosocial impact.
Derosier R, Farber RS. Polisher Research Institute, Madlyn and Leonard Abramson Center for Jewish Life, 1425 Horsham Road, North Wales, PA 19454, USA. Work. 2005;25(2):125-34.
The purpose of this study was to gather data concerning the psychosocial (quality of life) impact of speech recognition software on individuals with physical disabilities and to identify how satisfied these individuals were with this software as a computer access method. Two standardized questionnaires, the Psychosocial Impact of Assistive Devices Scale (PIADS) and the Quebec User Evaluation of Satisfaction with assistive Technology (QUEST) were administered to ten participants with physical disabilities who received speech recognition software following an assistive technology evaluation. The results of this study indicated that 90% of the participants were quite satisfied with speech recognition software as an assistive device and that the software had a somewhat positive psychosocial impact on their lives. Four themes emerged concerning what the participants liked most about the software: 1) the software provided a method of access when they were not previously accessing a computer, 2) the software increased independence, 3) the software made computer use more efficient, and 4) the software provided a choice or flexibility in computer access. Although this study demonstrated that these speech recognition software users are generally satisfied with the software and it has had a positive impact on their life, it also suggests that there is a need to examine the role of training on satisfaction and successful use of the software.