Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Feel my Speech: Automatic Speech Emotion Conversion for Tangible, Haptic, or Proxemic Interaction Design

Ilhan Aslan

TL;DR

The paper addresses turning speech emotion representations into physical sensations to enable tangible, social, and proxemic interactions. It introduces a three-part SEC framework and a starter kit that links SER outputs to physical displays (e.g., Arduino vibrotactile motors) and visualization tools via a CLI, leveraging the AffectToolbox for affect sensing. It highlights design opportunities in animal–computer interaction, proxemic interaction, and somaesthetic design, plus potential artistic applications like the Affective Bar Piano. By reducing technical barriers and reframing emotions as design material, the work broadens the practical impact of SER for interdisciplinary interaction design and artistic research, situating SEC within the transformer-driven and foundation-model–influenced evolution of affective computing.

Abstract

Innovations in interaction design are increasingly driven by progress in machine learning fields. Automatic speech emotion recognition (SER) is such an example field on the rise, creating well performing models, which typically take as input a speech audio sample and provide as output digital labels or values describing the human emotion(s) embedded in the speech audio sample. Such labels and values are only abstract representations of the felt or expressed emotions, making it challenging to analyse them as experiences and work with them as design material for physical interactions, including tangible, haptic, or proxemic interactions. This paper argues that both the analysis of emotions and their use in interaction designs would benefit from alternative physical representations, which can be directly felt and socially communicated as bodily sensations or spatial behaviours. To this end, a method is described and a starter kit for speech emotion conversion is provided. Furthermore, opportunities of speech emotion conversion for new interaction designs are introduced, such as for interacting with animals or robots.

Feel my Speech: Automatic Speech Emotion Conversion for Tangible, Haptic, or Proxemic Interaction Design

TL;DR

The paper addresses turning speech emotion representations into physical sensations to enable tangible, social, and proxemic interactions. It introduces a three-part SEC framework and a starter kit that links SER outputs to physical displays (e.g., Arduino vibrotactile motors) and visualization tools via a CLI, leveraging the AffectToolbox for affect sensing. It highlights design opportunities in animal–computer interaction, proxemic interaction, and somaesthetic design, plus potential artistic applications like the Affective Bar Piano. By reducing technical barriers and reframing emotions as design material, the work broadens the practical impact of SER for interdisciplinary interaction design and artistic research, situating SEC within the transformer-driven and foundation-model–influenced evolution of affective computing.

Abstract

Innovations in interaction design are increasingly driven by progress in machine learning fields. Automatic speech emotion recognition (SER) is such an example field on the rise, creating well performing models, which typically take as input a speech audio sample and provide as output digital labels or values describing the human emotion(s) embedded in the speech audio sample. Such labels and values are only abstract representations of the felt or expressed emotions, making it challenging to analyse them as experiences and work with them as design material for physical interactions, including tangible, haptic, or proxemic interactions. This paper argues that both the analysis of emotions and their use in interaction designs would benefit from alternative physical representations, which can be directly felt and socially communicated as bodily sensations or spatial behaviours. To this end, a method is described and a starter kit for speech emotion conversion is provided. Furthermore, opportunities of speech emotion conversion for new interaction designs are introduced, such as for interacting with animals or robots.

Paper Structure

This paper contains 11 sections, 2 figures.

Table of Contents

  1. Introduction
  2. "Physical Emotion" Generation from Speech
  3. Speech Emotion Recognition (SER)
  4. Speech Emotion Conversion (SEC) Kit
  5. Realizing affect-sensing
  6. Computational Generation of "Physical" Emotions
  7. Design Opportunities created by SEC
  8. SEC for Animal-Computer and Animal-Human Interaction
  9. SEC for Proxemic Interaction
  10. SEC for Soemasthetic Interaction Design and Artistic Research
  11. Conclusion

Figures (2)

  • Figure 1: Overview of the starter kit. The main part of the kit is provided as a command line tool with a command line interface to easily modify options provided for the processing of the SER task and dealing with specifics of input and output. This part is typically running on a PC. The second part of the toolkit enables exemplary ways to display speech emotions as tangible or haptic sensations with the Arduino IDE.
  • Figure 2: Screenshot of the AffectToolbox mertes2024affecttoolbox GUI, with modules selected to perform SER only.