WarpNet: Self-Organizing Time Warping
نویسنده
چکیده
Motorola, Phoenix Corporate Research Laboratories, 2100 East Elliot Road, MD EL508, Tempe, AZ 85284, USA tel: (602)413-4129, fax: (602)413-7281, email: [email protected] Abstract We describe “WarpNet”, a time-warping algorithm for speech recognition based on neural nets. WarpNet is a self-organizing matching mechanism intended to replace dynamic programming (Dynamic Time Warping or Viterbi-algorithms). The concept is based on elastic models that self-organize in the spirit of Self Organizing Maps and Optimizing Maps [9] so as to minimize the the sum of distances over two utterances, or any time series. We show that both the recognition accuracy, and the computing time of WarpNet are comparable to the optimal matching method, dynamic programming.
منابع مشابه
Motive Identification in 22 Folksong Corpora Using Dynamic Time Warping and Self Organizing Maps
A system for automatic motive identification of large folksong corpora is described in this article. The method is based on a dynamic time warping algorithm determining inherent repeating elements of the melodies and a self-organizing map that learns the most typical motive contours. Using this system, the typical motive collections of 22 cultures in Eurasia have been determined, and another gr...
متن کاملAnalyzing bird song syllables on the Self-Organizing Map
In this work, we present two methods for analyzing the syllables of the bird song on the Self-Organizing Map (SOM). Dynamic time warping is used for computing the distances between the data sequences. In the first method, the pairwise distances are first computed between the data sequences and each row of the distance matrix is then considered as a feature vector. The conventional SOM with fixe...
متن کاملAutomatic Registration of Complex Images Using a Self Organizing Neural System 1
| We present a system for automatic mapping of complex gray-scale images onto each other. The system includes a Neocognitron-like structure for hierarchical feature extraction, a 3D Self Organising Map to determine feature classes for unsupervised training, and algorithmic methods for landmark correspondence and image warping. We present results showing successful registration of MRI brain scan...
متن کاملTime Warping Techniques in Clustering Time Series
The problem of obtaining an accurate forecast is becoming more and more significant as the production possibilities and technologies evolve. The accuracy of a forecast mainly depends on the dataset used for forecasting as well as on the methods employed. This study is devoted to the time series analysis. A set of known methods and techniques are used for analysing the time series; one of them i...
متن کاملSpatio-Temporal Self-Organizing Feature Maps
Thus far, the success of capturing and classifying temporal information with neural networks has been limited. Our methodology adds a spatio-temporal coupling to the Self-Organized Feature Map (SOFM) which creates temporally and spatially localized neighborhoods in the map. The spatio-temporal coupling is based on traveling waves of activity starting at each winning node which are naturally att...
متن کامل