Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Why So Serious? Exploring Timely Humorous Comments in AAC Through AI-Powered Interfaces

Tobias Weinberg, Kowe Kadoma, Ricardo E. Gonzalez Penuela, Stephanie Valencia, Thijs Roumen

TL;DR

The paper addresses how AAC users can deliver timely humorous comments in ongoing conversations by introducing AI-powered interfaces with varying levels of automation. Through seven qualitative interviews and a five-participant user study, it reveals a critical tension between agency and efficiency, showing that users are sometimes willing to trade control for speed in time-sensitive humor. Four interface designs (Context Bubble Selection, Keywords, Wizard, and Full-auto) explore this trade-off, with results highlighting that Context Bubble Selection best preserves user intent while Full-auto maximizes timeliness. The work offers design recommendations for adaptive, privacy-conscious AAC systems that balance expressivity with efficiency, aiming to empower users to express humor authentically in real-time interactions.

Abstract

People with disabilities that affect their speech may use speech-generating devices (SGD), commonly referred to as Augmentative and Alternative Communication (AAC) technology. This technology enables practical conversation; however, delivering expressive and timely comments remains challenging. This paper explores how to extend AAC technology to support a subset of humorous expressions: delivering timely humorous comments -- witty remarks -- through AI-powered interfaces. To understand the role of humor in AAC and the challenges and experiences of delivering humor with AAC, we conducted seven qualitative interviews with AAC users. Based on these insights and the lead author's firsthand experience as an AAC user, we designed four AI-powered interfaces to assist in delivering well-timed humorous comments during ongoing conversations. Our user study with five AAC users found that when timing is critical (e.g., delivering a humorous comment), AAC users are willing to trade agency for efficiency contrasting prior research where they hesitated to delegate decision-making to AI. We conclude by discussing the trade-off between agency and efficiency in AI-powered interfaces, how AI can shape user intentions, and offer design recommendations for AI-powered AAC interfaces.

Why So Serious? Exploring Timely Humorous Comments in AAC Through AI-Powered Interfaces

TL;DR

The paper addresses how AAC users can deliver timely humorous comments in ongoing conversations by introducing AI-powered interfaces with varying levels of automation. Through seven qualitative interviews and a five-participant user study, it reveals a critical tension between agency and efficiency, showing that users are sometimes willing to trade control for speed in time-sensitive humor. Four interface designs (Context Bubble Selection, Keywords, Wizard, and Full-auto) explore this trade-off, with results highlighting that Context Bubble Selection best preserves user intent while Full-auto maximizes timeliness. The work offers design recommendations for adaptive, privacy-conscious AAC systems that balance expressivity with efficiency, aiming to empower users to express humor authentically in real-time interactions.

Abstract

People with disabilities that affect their speech may use speech-generating devices (SGD), commonly referred to as Augmentative and Alternative Communication (AAC) technology. This technology enables practical conversation; however, delivering expressive and timely comments remains challenging. This paper explores how to extend AAC technology to support a subset of humorous expressions: delivering timely humorous comments -- witty remarks -- through AI-powered interfaces. To understand the role of humor in AAC and the challenges and experiences of delivering humor with AAC, we conducted seven qualitative interviews with AAC users. Based on these insights and the lead author's firsthand experience as an AAC user, we designed four AI-powered interfaces to assist in delivering well-timed humorous comments during ongoing conversations. Our user study with five AAC users found that when timing is critical (e.g., delivering a humorous comment), AAC users are willing to trade agency for efficiency contrasting prior research where they hesitated to delegate decision-making to AI. We conclude by discussing the trade-off between agency and efficiency in AI-powered interfaces, how AI can shape user intentions, and offer design recommendations for AI-powered AAC interfaces.

Paper Structure

This paper contains 45 sections, 10 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Positionality Statement
  3. Contribution, Benefits, and Limitations
  4. Related Work
  5. Challenges in AAC: Timing, Agency, and Expression
  6. Trade-offs in AI Writing Support
  7. Computational Humor and Creative Expression Tools
  8. AAC & Comedy
  9. Qualitative Interviews: understanding use and challenges of AAC technology to create humorous comments
  10. Participants
  11. Structure
  12. Data and Analysis
  13. Results
  14. Reliance on AAC: "Communication is wild and crazy, but it is also the window to my world"
  15. Current Use and Experience of Humor with AAC
  16. ...and 30 more sections

Figures (10)

  • Figure 1: Overview of an example AAC interface to create humorous comments (Keywords interface). Screen 1: (a) Joke Mode (b) play text through TTS. Screen 2: (c) keywords extracted from the conversation (d) input field to type the starter of the joke (e) dynamically updated suggestion (f) get new suggestion
  • Figure 2: Interface 1--Context Bubble Selection features (a) dynamically updated suggestions (b) context selection from the transcript (c) new suggestions (d) exit joke mode (e) keywords extracted from the selected bubbles (f) output through TTS.
  • Figure 3: Interface 2 - Keywords, features (a) keywords extracted from the conversation (b) transcript from the conversation (c) dynamically updated suggestion (d) exit joke mode (e) get new suggestion (f) play text through TTS.
  • Figure 4: Interface 3--Wizard features (a) keywords from the conversation (b) association from the keywords (c) suggestions (d) exit joke mode (e) refresh suggestions
  • Figure 5: Interface 4 - Full-auto: features (a) transcript from the conversation (b) manual input text (optional) (c) best suggestion based on the conversation (d) exit joke mode (e) get new suggestion (f) play text through TTS.
  • ...and 5 more figures