Detecting Topic-Oriented Speaker Stance in Conversational Speech
- Submitting institution
-
University of Edinburgh
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 110910536
- Type
- E - Conference contribution
- DOI
-
10.21437/Interspeech.2019-2632
- Title of conference / published proceedings
- Proceedings of Interspeech 2019
- First page
- 46
- Volume
- -
- Issue
- -
- ISSN
- 1990-9772
- Open access status
- Compliant
- Month of publication
- September
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
5
- Research group(s)
-
D - Language, Interaction and Robotics
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This impactful paper, demonstrates for the first time that it is possible to automatically detect speaker stance in spoken dialogue using generic acoustic and lexical features. A speaker’s stance can be used to understand whether a conversation is going well (or not) from each speaker’s perspective and is particularly useful for assistive robots in building long-term engagement strategies with humans, making their support services more personalized and engaging. Early results from this work led to two grants from Toyota Europe who are planning to use it in their personal robot systems (contact: Senior Researcher).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -