Wearable and assistive speech-to-text technologies for people living with dementia: a scoping review of the evidence base

Many people living with dementia find it hard to follow spoken conversation, not because they cannot hear, but because spoken words disappear too quickly to take in. This difficulty is especially pronounced on the telephone, when the speaker is in another room, or when their face cannot be seen, and simply increasing volume does not help. Over time this can lead to missed information, embarrassment, and withdrawal from conversations and phone calls altogether, a pattern distinct from hearing loss and currently under-represented in assistive technology research.A growing consumer market of live-captioning wearables (smart glasses and similar devices) has emerged to support people with hearing loss, converting nearby speech into real-time, reviewable text. However, this market is designed around hearing loss as the presenting difficulty, and it is not yet established whether such devices, or the underlying automatic speech recognition (ASR) they depend on, perform adequately or are acceptable for people living with dementia specifically. Recent evidence indicates that ASR systems show significantly reduced accuracy for speech produced by adults with cognitive impairment, and that dementia-related speech differences (extended pauses, word-finding difficulty, disfluency) are not well handled by systems trained primarily on typical speech (Cohn et al., 2026). This raises a direct question for the present project: whether existing speech-to-text and captioning technologies can be adapted, or must be purpose-built, to work for this population.Existing reviews in this space have tended to focus either on assistive technology for dementia broadly, without isolating spoken-conversation comprehension as a distinct problem (e.g. Martin and Jones, 2025), or on captioning and hearing-assistive wearables for hearing loss, without reference to dementia or cognitive impairment at all. A gap exists in mapping what is known specifically about technologies that capture, convert, and allow review of spoken speech for people living with dementia: what has been built or tested, how people living with dementia have been involved in its design, what is known about its feasibility and acceptability, and how equity considerations have (or have not) been addressed.This scoping review maps the existing evidence on wearable and assistive speech-to-text technologies for people living with dementia. The review establishes the evidence base for the unmet need, positions Quo Verba against existing technologies, and directly feeds Work Package 2 (technology assessment and architecture selection).

Authors

Publication Details

Journal
Figshare
Published
2026-11-18
DOI
https://doi.org/10.6084/m9.figshare.35049628
Primary Topic
Assistive Technology in Communication and Mobility
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Wearable and assistive speech-to-text technologies for people living with dementia: a scoping review of the evidence base

Frank Ho‐yin Lai
Figshare
Assistive Technology in Communication and Mobility
article

Wearable and assistive speech-to-text technologies for people living with dementia: a scoping review of the evidence base

Frank Ho‐yin Lai
article en

Abstract

Many people living with dementia find it hard to follow spoken conversation, not because they cannot hear, but because spoken words disappear too quickly to take in. This difficulty is especially pronounced on the telephone, when the speaker is in another room, or when their face cannot be seen, and simply increasing volume does not help. Over time this can lead to missed information, embarrassment, and withdrawal from conversations and phone calls altogether, a pattern distinct from hearing loss and currently under-represented in assistive technology research.A growing consumer market of live-captioning wearables (smart glasses and similar devices) has emerged to support people with hearing loss, converting nearby speech into real-time, reviewable text. However, this market is designed around hearing loss as the presenting difficulty, and it is not yet established whether such devices, or the underlying automatic speech recognition (ASR) they depend on, perform adequately or are acceptable for people living with dementia specifically. Recent evidence indicates that ASR systems show significantly reduced accuracy for speech produced by adults with cognitive impairment, and that dementia-related speech differences (extended pauses, word-finding difficulty, disfluency) are not well handled by systems trained primarily on typical speech (Cohn et al., 2026). This raises a direct question for the present project: whether existing speech-to-text and captioning technologies can be adapted, or must be purpose-built, to work for this population.Existing reviews in this space have tended to focus either on assistive technology for dementia broadly, without isolating spoken-conversation comprehension as a distinct problem (e.g. Martin and Jones, 2025), or on captioning and hearing-assistive wearables for hearing loss, without reference to dementia or cognitive impairment at all. A gap exists in mapping what is known specifically about technologies that capture, convert, and allow review of spoken speech for people living with dementia: what has been built or tested, how people living with dementia have been involved in its design, what is known about its feasibility and acceptability, and how equity considerations have (or have not) been addressed.This scoping review maps the existing evidence on wearable and assistive speech-to-text technologies for people living with dementia. The review establishes the evidence base for the unmet need, positions Quo Verba against existing technologies, and directly feeds Work Package 2 (technology assessment and architecture selection).

Figshare
Good health and well-being
Assistive Technology in Communication and Mobility
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon