Automatic role recognition based on conversational and prosodic behaviour

This paper proposes an approach for the automatic recognition of roles in settings like news and talk-shows, where roles correspond to specific functions like Anchorman, Guest or Interview Participant. The approach is based on purely nonverbal vocal behavioral cues, including who talks when and how much (turn-taking behavior), and statistical properties of pitch, formants, energy and speaking rate (prosodic behavior). The experiments have been performed over a corpus of around 50 hours of broadcast material and the accuracy, percentage of time correctly labeled in terms of role, is higher than 85%. Both turn-taking and prosodic behavior lead to satisfactory results, but their combination does not lead to statistically significant changes of performance. To the best of our knowledge, this is the first attempt to use prosodic features in a role recognition experiment. Categories and Subject Descriptors: H.3.1 [Content Analysis and Indexing].General Terms: Experimentation.

• Type: Conference
• Year: 2010
• Where: MM
• Authors: Hugues Salamin, Alessandro Vinciarelli, Khiet Truong, Gelareh Mohammadi
• Keywords: Automatic Recognition, MM 2010, Multimedia, Prosodic Behavior, Vocal Behavioral Cues