Article

June 2015

Dysarthria and Speech Technology

Video - June 2015

Dysarthria and Speech Technology

Dysarthria is defined as a difficulty in speaking which can be caused by a number of different disorders such as strokes, brain injuries, brain tumors, Parkinson’s disease and multiple sclerosis.

Parkinson’s disease (PD), in particular, is estimated to affect approximately 1 million people in the US alone and more than 7 million people worldwide. These numbers are generally considered as an underestimation of the true incidence rate since this disease has a direct relation with aging. So, as the population is growing older, this disease is becoming more and more prevalent. The symptoms of PD may include tremors, slowed movement, rigid muscles and speech changes. Reliable PD diagnosis is difficult and researchers have shown that speech can be used as a decision support to discriminate PD symptoms.

Slight changes in the voice such as tremors and reduced amplitude can indicate impairments in the vocal tracts which can be considered as early signs of PD. These symptoms may happen up to five years before other prominent symptoms appear. Using speech technology, these impairments can be identified well in advance. There are some apps already developed to monitor PD, such as mPower which uses different sensors on the phone to monitor the progress of PD. However, speech is not utilized as a part of this app which definitely can help to better monitor the progress of PD. Rather than monitoring the advancement of PD, elderly people who are more vulnerable to PD can also benefit from this technology by monitoring their speech regularly. In case of the presence of any early signs of PD, such patients can be referred to more specific clinical tests. Examining voice using a cell phone can be done by the patient himself without requiring access to any other resources (such as GPs or clinical tests) and this has some extra benefit for underdeveloped countries.

Dysarthric speech can also be enhanced by speech technology to make it more understandable. Earlier, speech synthesizers and text to speech convertors were used by several dysarthric speakers. However, the advances in speech technology made it possible to directly enhance dysarthric speech. Researchers have shown that energy modifications and enhancing the formant of dysarthric speech can greatly increase its comprehensibility. With computing advances, the enhancement processing can be done in real time. Thus, while a patient with dysarthric speech is speaking, speech technology based devices can enhance the speech signal aiding in better perception.

In the next post we will talk more about other potential applications of the speech technology in the health sector.

References:
http://www.nlm.nih.gov/medlineplus/ency/article/007470.htm
http://www.mayoclinic.org/diseases-conditions/parkinsons-disease/basics/symptoms/con-20028488
http://www.urmc.rochester.edu/news/story/index.cfm?id=4268
http://parkinsonmpower.org/
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5646752&tag=1

About the author

Never miss an insight
Get insights delivered right to your inbox

Never miss an insight

Get insights delivered right to your inbox
By signing up to receive our newsletter, you agree to our Terms of Use and Privacy Policy. You can unsubscribe at anytime.