Category: Allied Healthcare Professionals
Objective: The aim of the present study was to examine the Automatic Speech Recognition (ASR) technology as a useful approach to enhance the speech intelligibility of a Parkinson’s disease (PD) patient.
Background: Speech disorder is a common difficulty in PD, affecting communication and quality of life. Early referral to speech and language therapy is recommended for the management of dysarthria. Furthermore, several training approaches have been described to sustain speech abilities in this population, and ASR systems have been recently adopted for the objective analysis of dysarthric speech. However, few studies have explored the use of this technology for the treatment of dysarthria despite its potential usefulness in providing immediate feedback of speech intelligibility during the conversation.
Method: A patient at the late stage of PD underwent a year of speech training at home, two sessions per week. The evaluation of dysarthria, including acoustic analysis of speech, the Robertson Profile and self-questionnaires on speech and quality of life, have been administered before the training, at 6-months follow-up, and at the end of the year training.
The patient has been trained to use an ASR system on daily basis with an increased difficulty of speech tasks in addition to the standard training.
Results: The patient demonstrate increased intelligibility of speech resulting in the ability to correctly dictate short paragraphs of about 50 words using ASR with a level of word correctness of 83% in comparison to the 65% detected at the beginning of the training. The patient reported an improvement of subjective perception of speech and quality of life parameters.
Conclusion: The ASR technology should be further investigated for the treatment of dysarthria in a larger sample as an additional tool for proving continuous feedback and improving speech intelligibility in PD.
To cite this abstract in AMA style:
A. Nuzzi. Automatic Speech Recognition for speech therapy in a Parkinson’s disease patient [abstract]. Mov Disord. 2022; 37 (suppl 2). https://www.mdsabstracts.org/abstract/automatic-speech-recognition-for-speech-therapy-in-a-parkinsons-disease-patient/. Accessed November 21, 2024.« Back to 2022 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/automatic-speech-recognition-for-speech-therapy-in-a-parkinsons-disease-patient/