Smart speakers help people with intellectual disabilities speak more clearly

Commonly available smart speakers appear to be beneficial for adults with intellectual disabilities (IDs), demonstrating marked improvements in their speech intelligibility, according to a new study.

“We used a semi-randomized controlled trial to show that using a smart speaker for about 12 weeks could improve ratings of speech intelligibility in adults with ID for both smart speaker-related phrases and unrelated words,” the researchers said. “These initial findings suggest that smart speaker technology could be a novel, and inclusive, route to improving speech intelligibility in vulnerable populations.”

Adults with ID were assigned to either an intervention group (n=21), who were given smart speaker devices, or to a control group (n=22). At baseline and after about 12 weeks, participants were recorded saying phrases related to device use, or totally unrelated words, and naïve listeners were asked to rate the intelligibility according to a seven-point scale.

In terms of device-related phrases (e.g., “play some music”), multilevel linear mixed effect models demonstrated a significant interaction between the time and group factors (p<0.001). Posthoc comparisons showed that participants given smart speakers had greater improvements in intelligibility ratings (mean, 0.36; p<0.001) than their control comparators (mean, 0.11; p<0.01). [Int J Lang Commun Disord 2021;doi:10.1111/1460-6984.12615]

Of note, intelligibility ratings at baseline were numerically different between intervention and control groups, but not significantly so (p=0.44). Moreover, the researchers detected no significant main effect of group (p=0.31), while such an effect was positively recorded for time (p<0.001).

“Note that time was within-subject, while group was between subject and the ID population had wide heterogeneity. We could not pairwise control for speech intelligibility ratings; households of participants were randomly allocated to either group,” the researchers said. But despite the heterogeneity, the main effect of time remained significant.

In turn, they sought to replicate their initial findings, using a second identical model to test whether the smart speakers could elicit a similar effect on words unrelated to device use.

The interaction between the time and group factors was also significant (p<0.01), and posthoc analysis again showed that intelligibility ratings increased with greater magnitude in the intervention (mean, 0.13; p<0.001) than in the control (mean, 0.02; p=0.4) group. The main effect of time remained significant (p<0.001) while that of group was not (p=0.06).

“Speech improvement was found for phrases that were related to device use, and the effect was replicated, although a little smaller, for unrelated words, thus going beyond specific practiced phrases,” the researchers said.

“This finding is particularly significant given that the smart speakers were not actively provided as a speech training tool, but rather to improve quality of life with entertainment functions,” suggesting that the gains in intelligibility were due to unstructured use of the smart speakers, they added.

“These findings have potential relevance for other groups experiencing speech and language difficulties and could viably offer a simple-to-deliver, affordable route to supplement speech and language therapy at scale,” the researchers said. “Future research should further investigate the mechanisms behind these improvements.”