Prediction of a hotspot pattern in keyword search results.
Computer Speech & Language(2018)
摘要
•Hotspots of keyword search detections over speech recognition output are identified.•Keyword hotspots are modeled as Hawkes process to automatically label data.•Prosodic features are used to predict hotspots for word-sized time intervals.•Two challenges, class imbalance and disparity of training and test data, are addressed.•A novel data selection method with good generalization properties is proposed.
更多查看译文
关键词
Self-exciting point process,Data selection,Keyword search,Speech retrieval
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要