Article
Delta Coefficients based Bacterial Foraging Optimization with RBF Neural Network for Speaker Recognition
Speaker recognition is essential in the field of authentication and surveillance to validate the identity of the user using the extracted features of the audio speech signal. A novel Delta Coefficients-based Bacterial Foraging Optimization (BFO) with Radial Basis Function (RBF) for identification of exact speaker is proposed in this paper. The speaker recognition function can be done with two modules, namely Training and Testing. Initially throughout the training step, the Delta Coefficients of each speech sample are derived by pre-processing the audio speech signal. Then the features are classified towards the target speaker using RBFNN. Bottleneck features for abstract representation are extracted in RBFNN and the dimensionality by Principal Component Analysis (PCA) is further decreased. To improve the classification accuracy, the features are optimized with Bacterial Foraging optimization (BFO) algorithm. Finally, the probability score for each speaker is generated to identify the speaker. The proposed Delta Coefficients based BFO using Bottleneck features (DBFOB) method is evaluated with TIMIT data corpus. The improved performance is presented with regard to Equal Error Rate (EER).
Full Text Attachment
