The value of visualizing tone of voice

Abstract

Whilst most of us have an innate feeling for tone of voice, it is an elusive quality that even phoneticians struggle to describe with sufficient subtlety. For people who cannot speak themselves this can have particularly profound repercussions. Augmentative communication often involves text-to-speech, a technology that only supports a basic choice of prosody based on punctuation. Given how inherently difficult it is to talk about more nuanced tone of voice, there is a risk that its absence from current devices goes unremarked and unchallenged. Looking ahead optimistically to more expressive communication aids, their design will need to involve more subtle interactions with tone of voice–interactions that the people using them can understand and engage with. Interaction design can play a role in making tone of voice visible, tangible, and accessible. Two projects that have already catalysed interdisciplinary debate in this area, Six Speaking Chairs and Speech Hedge, are introduced together with responses. A broader role for design is advocated, as a means to opening up speech technology research to a wider range of disciplinary perspectives, and also to the contributions and influence of people who use it in their everyday lives.

Key words::

• Assistive technology
• augmentative and alternative communication
• phonetics
• prosody
• interaction design
• intonation
• speech-generating devices
• speech technology
• text-to-speech
• voice output communication aids

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

The authors are grateful to the EPSRC and the School of Computing, University of Dundee, for the funding of Andrew Cook's PhD Scholarship. We would also like to thank ACE North, Centre for Speech Technology Research, University of Edinburgh, Communication Matters and AAC–RERC for their generosity in supporting our travel and/or engaging with our research.