Dimensionality reduction for using high-order n-grams in SVM-based phonotactic language recognition
Mikel Peñagarikano, Amparo Varona, Luis Javier Rodríguez-Fuentes, Germán Bordel · 2011
SVM-based phonotactic language recognition is state-of-the-art technology. However, due to computational bounds, phonotac-tic information is usually limited to low-order phone n-grams (up to n = 3). In a previous work, we proposed a feature selec-tion algorithm, based on n-gram frequencies, which allowed us work successfully with high-order n-grams on the NIST 2007 LRE database. In this work, we use two feature projection met-hods for dimensionality reduction of feature spaces including up to 4-grams: Principal Component Analysis (PCA) and Ran-dom Projection. These methods allow us to attain competitive performance even for small feature sets (e.g. of size 500). Sys-tems were built by means of open software (BUT phone deco-ders, HTK, SRILM, LIBLINEAR and FoCal) and experiments were carried out on the NIST 2009 LRE database. Best perfor-mance was attained by using the feature selection algorithm to get around 11500 features: 1.93 % EER and CLLR = 0.413. When considering smaller sets of features, PCA provided best performance. For instance, using PCA to get a 500-dimensional feature subspace yielded 2.15 % EER and CLLR = 0.457 (25% improvement with regard to using feature selection).