MFCC Based Speech Retrieval

Abstract

This paper presents an approach for speech retrieval. The feature being used in this approach is MFCC. This approach does not use any phoneme recognizer or Speech to text tool hence it can be used for other languages as well leads to the problem of speech retrieval (SR). This method retrieves ranked audio files containing spoken text in response to a given speech query. In this paper indexing methods are described which represent the contents of the spoken documents. The indexing methods, which are based on the output of phoneme recognizer, take account of speech recognition errors. While in this paper, speech documents are directly compared with the speech query based on MFCC. Thus, reduced the overhead of conversion from speech to text.