Soft vector quantization and the EM algorithm
Neural Networks
A Middleware Infrastructure for Active Spaces
IEEE Pervasive Computing
A study of results overlap and uniqueness among major web search engines
Information Processing and Management: an International Journal
Hadoop: The Definitive Guide
Classification of audio signals using AANN and GMM
Applied Soft Computing
Tracking of Nonstationary Noise Based on Data-Driven Recursive Noise Power Estimation
IEEE Transactions on Audio, Speech, and Language Processing
| Hi-index | 0.00 |
Smart city means an intelligent city called post-ubiquitous city. However, to be a smart city, there are some issues to consider carefully. The first is the security issues having platform and intelligent surveillance function and analysis function as well as inference. The second is the governance issues between government and cities which are developing. And the third is the service issues to be realized in the cities. In this paper, we are mainly concentrated on the first security issue. Here we propose a speech recognition technology in emenrgency situation for secure cities. For the emergency detection in general CCTVs(closed circuit television) environment of our daily life, the monitoring by only images through CCTVs information occurs some problems especially in emergency state. Therefore for detecting emergency state dynamically through CCTVs as well as resolving some problems, we propose a detection and recognition method for emergency and non-emergency speech by Gaussian Mixture Models(GMM). The proposed method determines whether input speech is emergency or non-emergency speech by global GMM firstly. If this is an emergency speech, then local GMM is performed secondly to classify the type of emergency speech. The proposed method is tested and verified by emergency and non-emergency speeches in various environmental conditions. Also, we discuss about the platform issues having analysis and inference function of big data in smart city.