作者: A. Pentland , B. Clarkson , N. Sawhney
DOI:
关键词:
摘要: We describe a system for obtaining environmental context through audio applications and user interfaces. are interested in identifying specific auditory events such as speakers, cars, shutting doors, scenes the office, supermarket, or busy street. Our goal is to construct that real-time robust variety of real-world environments. The current detects classifies using HMM framework. design around an adaptive structure relies on unsupervised training segmentation sound scenes.