Speech Recognition using MFCC
Siwat Suksri · 2015
This paper describes an approach of speech recognition by using the Mel-Scale Frequency Cepstral Coefficients (MFCC) extracted from speech signal of spoken words. Principal Component Analysis is employed as the supplement in feature dimensional reduction state, prior to training and testing speech samples via Maximum Likelihood Classifier (ML) and Support Vector Machine (SVM). Based on experimental database of total 40 times of speaking words collected under acoustically controlled room, the sixteen-ordered MFCC extracts have shown the improvement in recognition rates significantly when training the SVM with more MFCC samples by randomly selected from database, compared with the ML. Keywords—Speech Signal, MFCC, SVM, ML I. INTRODUCTION PEECH recognition is the process of automatically recognizing the spoken words of person based on information in speech signal. Recognition technique makes it possible to the speaker's voice to be used in verifying their identity and control access to services such as voice dialing, banking by telephone, telephone shopping, database access services, information service, voice mail, security control for the confidential information areas, and remote access to computers. The acoustical parameters of spoken signal used in recognition tasks have been popularly studied and investigated, and being able to be categorized into two types of processing domain: First group is spectral based parameters and another is dynamic time series. The most popular spectral based parameter used in recognition approach is the Mel Frequency Cepstral Coefficients called MFCC (2,3). Due to its advantage of less complexity in implementation of feature extraction algorithm, only sixteen coefficients of MFCC corresponding to the Mel scale frequencies of speech Cepstrum are extracted from spoken word samples in database. All extracted MFCC samples are then statistically analyzed for principal components, at least two dimensions minimally required in further recognition performance evaluation.