Exploring use cases and user perception of a proactive voice assistant in automated vehicles

Mathis, Lesley-Ann; Werner, Kathrin; Widlroither, Harald · 2023 · Crossref

DOI: 10.54941/ahfe1003803

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the potential use cases and user perception of proactive voice assistants (VA) in automated vehicles. As driving automation shifts the user’s role from driver to passenger, there is an opportunity for VAs to initiate interactions rather than merely responding to commands. However, research on proactive VA behaviors during automated rides, particularly for non-driving-related activities, remains limited. The authors aim to derive and evaluate specific use cases for a proactive VA, focusing on how users perceive these interactions in a simulated SAE Level 4 automated driving environment. The researchers employed a multi-step methodology involving expert workshops, an online survey, and a driving simulator study. First, five experts brainstormed potential use cases based on a specific persona ("Matthias," a sales employee) and a business ride journey. These ideas were clustered into five categories: office work, personalization, entertainment, knowledge, and well-being. Second, 80 participants rated 24 distinct use cases in an online survey using scenario descriptions and Likert-scale items assessing usefulness, efficiency, comfort, and other factors. Finally, the top 10 use cases were tested in a Wizard-of-Oz driving simulator study with 32 participants. During a 40-minute automated ride, participants experienced proactive prompts every 2–6 minutes. Their qualitative feedback was collected via Retrospective Think-Aloud (RTA) interviews and analyzed using qualitative content analysis. The results indicate that office-work-related use cases, such as meeting preparation and to-do list management, received the highest ratings for usefulness and efficiency. Well-being-related features, like movement exercises, were also favorably rated for comfort and excitement. In contrast, entertainment-focused use cases, including gaming and small talk, received lower approval. Qualitative analysis revealed that users generally found the proactive VA pleasant and useful, particularly when it helped utilize travel time efficiently or supported their well-being. However, some users expressed concerns about losing control or feeling patronized, noting a preference for manual adjustment of certain settings. Participants emphasized the importance of adaptivity, requesting the ability to customize topics and disable specific proactive features, such as motion sickness warnings, if they did not apply to them. The study concludes that proactive VAs are most accepted when they provide clear benefits in efficiency and comfort, rather than serving purely as companions for entertainment. The findings highlight that user acceptance depends heavily on the system’s ability to adapt to individual preferences and routines, as well as the non-intrusive formulation of suggestions. The authors suggest that future research should focus on long-term adaptivity and real-world validation to address concerns regarding control and personalization. This work provides a foundational approach for deriving and evaluating proactive VA use cases, which can be adapted for different personas and application domains.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-09; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• generative ai voice assistants
• voice interaction