"Alexa, can you help me speak more clearly?"

Background: People with Parkinson’s disease have speech and voice difficulties that impact volume, clarity of speech and intelligibility. Domestic (commonly available) Voice assisted technology, like Alexa, poorly recognises speech difficulties and prompts people to change their speech to enable interaction with technology.

Aims: This study aims to identify and map available literature regarding the utilization of voice assisted technology to address speech and voice difficulties associated with Parkinson’s Disease and other conditions with related characteristics. We explore: (1) how well does voice assisted technology recognize speech difficulties or dysarthric speech, across different severities and perceptual voice categories and (2) how does using VAT impact intelligibility, clarity, or volume for people with dysarthria or speech difficulties?

Methods and Procedures: A review of available literature was guided by Arksey and O’Malley’s (2005) scoping review framework between September 2023 and November 2023. Five electronic databases were systematically searched: CINAHL Ultimate, Linguistic and Language Behavioural Abstracts (LLBA), Scopus, MEDLINE (Ovid) and ACM digital library. The Population, Exposure, and Outcome framework was used to develop inclusion and exclusion criteria. Results were screened, eligible papers identified and articles using out of the box voice assisted technology were included. Relevant data was extracted using an adapted version of the Joanna Briggs Institute extraction tool, and study outcomes were narratively reviewed. We report findings using the Preferred Reporting Items for Systematic Reviews and Meta-analysis extension for scoping reviews (PRISMA-ScR).

Results: Initial searches yielded 751 studies, and 656 remained following removal of duplicates, and were abstracts were screened. 52 studies were included for full text review and 47 articles were excluded. 5 studies reported on recognition rate of voice assisted technology or impact of using voice assisted technology on speech and were included for review. Studies were wide ranging geographically coming from 4 different countries, and varying speech difficulties to include dysarthria, stammering and reduced intelligibility. Results indicate out of the box voice assisted technology has limited applications for assessment in speech and language therapy. Despite this, it shows great promise as a tool for managing volume, intelligibility and clarity of speech, and has positive implications for activity, participation and wellbeing. Out of the box voice assisted technology may potentially clients with mild-moderate speech difficulties, as a home practice tool to promote generalisation and maintenance of speech gains following formal therapy.

Conclusions & Implications: This review provides a foundation for future co-design, development, and implementation of out of the box voice assisted technology as speech improvement facilitator. Future research should explore the views of people with Parkinson’s and speech and language therapists who are using voice assisted technology to understand usage. Evidence of effectiveness will depend upon future research evaluation.

What is already known on this subject:
Voice assisted technology poorly recognises dysarthric speech, which prompts some affected people to change their speech, speaking clearly and loudly, to interact with the technology. A review of literature is needed to understand if voice assisted technology can recognise or positively change speech and voice difficulties.

What this study adds:
Out of the box voice assisted technology shows potential as a management tool, particularly to improve intelligibility and volume. People with mild to moderate dysarthria, at earlier stages of disease progression would benefit most from using VAT, as an at home practice tool, for example following a formal Lee Silverman Voice Treatment. There is evidence that out of the box voice assisted technology has limited implications for assessment of speech difficulties.

Clinical Implications of this study
This review could inform clinical guidance to support speech and language therapists to use voice assisted technology as a clinical tool to manage speech and voice difficulties associated with Parkinson’s Disease