Table 1.
Algorithm 1: Dynamic Time Warping.
Figure 1.
Summary of the proposed one-against-all weighted dynamic time warping (OAWDTW) approach.
First, OAWDTW normalizes training MFCCs and testing MFCCs. Then, OAWDTW acquires the one-against-all index (OAI) of each time frame in training MFCCs. Third, OAWDTW performs dynamic time warping alignment between a normalized training MFCC and a normalized testing MFCC. Forth, OAWDTW applies OAIs of aligned time frames in normalized training MFCC to tune the final score.
Figure 2.
The illustration of MFCC normalization. In table header, ‘Fr’ represents time frame, and ‘Dim’ means dimension.
A MFCC is represented by a matrix. This matrix is constituted by
time frames where each time frame is represented by a
dimensional vector. Therefore, each dimension has
values, which will be normalized into the range between −1 and 1 after the MFCC normalization step.
Figure 3.
The illustration of one-against-all index (OAI) acquisition.
Suppose that there are three training MFCCs. We take the node of training MFCC
(bold and italic letter) as example. The
is acquired by using all of the distances among nodes in
and their aligned nodes. The
is acquired by using all the distance between node
in
and its aligned nodes.
Table 2.
Algorithm 2: Similarity Score of Normalized Training and Testing MFCC.
Figure 4.
The samples of clean and reverberant signals of audio “MinZhang”.
Table 3.
Recording Condition Description.
Table 4.
List of Recorded Names.
Table 5.
Accuracy (%) by using original DTW on different (non)normalized dimensional MFCC.
Table 6.
DTW accuracy (DTW Acc) and OAWDTW accuracy (OAWDTWAcc)(%) comparison, and OAWDTW relative reduction of error rate (OAWDTW RRER) (%) based on DTW.