Effect of Spoken Text on Text-Independent Speaker Recognition

Automatic speaker recognition (ASR) is a well-investigated area among the researcher of speech processing. Many of the factors that affect the recognition rate of an ASR system are addressed in the literature. Those factors are clean and noisy environments, cross channel, session variability, age and health of a speaker etc. One of the factors that still needs investigation is the effect of type of spoken text on speaker recognition (SR). In this paper, the developed ASR system is trained and tested with different types of text, to observe the effect of the type of text on SR performance. The text includes digits, words, sentences, and paragraphs. All other factors like session, environment, and recording channel are same for this study. The study concluded that careful design of the training and testing text can increase the recognition rate of an ASR system.