Employees’ acceptance of AI-based emotion analytics from speech on a group level in virtual meetings

Abstract

Detecting emotions in virtual meetings can be a demanding task for humans due to technological constraints of current videoconferencing systems. AI-based emotion analytics from speech is a promising technological option to provide helpful analyses for online business meetings, but might be perceived as scary as emotions are a sensitive topic. However, employees' acceptance of such a type of software is not understood properly yet as there are no offers on the market yet. To investigate potential users' intentions, we conducted a survey in 2021 as well as in 2023 with data from 470 employees in Germany with regard to a novel type of software analyzing emotions from speech on a group level in virtual meetings. We argue that employees' decision of acceptance bases on balancing different pros and cons for sticking with an existing behavior or using an emotion analytics software. A partial least squares (PLS) approach to structural equation modeling (SEM) was used to test the study's hypotheses. Our results show that attitude, perceived norms, perceived efficacy and perceived threats are significant predictors for the intention to accept such a software. At the same time, privacy concerns were only in 2023 a reason to reject the software. This paper contributes to a better understanding on why employees are willing to use emotion analytics software.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 19 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.