Skip to main content

Table 2 Influence of the number of differential coefficients with the HLDA transformation on phone recognition rates on the converted \(MFCC^*\) vectors of the Test part of FPSD database

From: A preliminary study on improving the recognition of esophageal speech using a hybrid system based on statistical voice conversion

36 monophone HMMs with 16 Gaussians per state \(+\) Bigram Accuracy (%) Correct (%)
Exp 1 : \(39 \ MFCC^*\) coefficients 63.48 68.58
Exp 2 : \(52 \ MFCC^*\) coefficients 61.78 67.36
Exp 3 : \(HLDA \ (52\rightarrow 39)\) 65.29 69.85