Frame distance array algorithm parameter tune-up for TIMIT corpus automatic speech segmentation | IEEE Conference Publication | IEEE Xplore

Frame distance array algorithm parameter tune-up for TIMIT corpus automatic speech segmentation


Abstract:

This work is related to unsupervised automatic speech segmentation. An experiment was carried out on the Frame Distance Array (FDA) algorithm with a main goal of the algo...Show More

Abstract:

This work is related to unsupervised automatic speech segmentation. An experiment was carried out on the Frame Distance Array (FDA) algorithm with a main goal of the algorithm parameter tune-up. The experiment was carried out by applying the algorithm on TIMIT corpus and by using MFCC as the speech signal features. The parameters tuned up in this work are the frame length, the frame increment, the number of test frames and the test frame step size. The best combination of values was chosen based on the observations on the detection rate, the miss rate and the false boundary rate. The best parameter tune-up found at 23 ms, 1.5 ms, 9 frames and 2 frames for the frame length, the frame increment, the number of test frames and the test frame step size respectively.
Date of Conference: 21-23 May 2015
Date Added to IEEE Xplore: 08 October 2015
ISBN Information:

ISSN Information:

Conference Location: Dekalb, IL, USA

References

References is not available for this document.