Role Recognition in Broadcast News using Bernoulli Distributions
ICME(2007)
摘要
This work presents an approach for the recognition of the roles played by speakers participating in radio broadcast n ews (e.g. anchorman or guest). The approach includes two main stages: the first is the split of the news recordings into sing le speaker segments using an unsupervised approach. The sec- ond is the application of Bernoulli Distributions for role m od- eling and recognition. The experiments are performed over a collection of 96 news bulletins (around 19 hours of material) and show that around 80 percent of the data time is labeled correctly in terms of role.
更多查看译文
关键词
radio broadcasting,speech processing,statistical distributions,Bernoulli distribution,radio broadcast news,role recognition,single speaker segment,speaker participation,unsupervised approach
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络