Inferring Stance from Prosody

Citation

Ward, N.G., Carlson, J.C., Fuentes, O., Castan, D., Shriberg, E.E., Tsiartas, A. (2017) Inferring Stance from Prosody. Proc. Interspeech 2017, 1447-1451, DOI: 10.21437/Interspeech.2017-159.

Abstract

Speech conveys many things beyond content, including aspects of stance and attitude that have not been much studied. Considering 14 aspects of stance as they occur in radio news stories, we investigated the extent to which they could be inferred from prosody.  By using time-spread prosodic features and by aggregating local estimates, many aspects of stance were at least somewhat predictable, with results significantly better than chance for many stance aspects, including, across English, Mandarin and Turkish, good, typical, local, background, new information, and relevant to a large group.

Index Terms: information retrieval, filtering, attitudes, sentiment, broadcast news


Read more from SRI

  • Banner and attendees at the IEEE Hard Tech Venture Summit

    Cultivating hard tech startups that scale

    IEEE’s Hard Tech Venture Summit convened innovators at SRI to refine strategies and build new networks.

  • Patient going into a MRI

    Bringing surgical tools inside the MRI

    Drawing on SRI’s unique innovation ecosystem, the startup Medical Devices Corner is seeking to improve cancer surgery by advancing MRI-safe teleoperation.

  • Christopher Mims and Susan Patrick

    PARC Forum: How to AI

    The Wall Street Journal tech columnist Christopher Mims and SRI Education’s Susan Patrick discuss how AI can strengthen human agency.