A Novel Technique for Speech Recognition and Visualization Based Mobile Application to Support Two-Way Communication between Deaf-Mute and Normal Peoples

Table 2

Details of a configuration file (config.txt).

Description	Parameters
Input Source File Format ()	SOURCEFORMAT = WAV
Output of Speech Sample	TARGETKIND = MFCC_0
Pre-emphasis Coefficient ()	PREEMCOEF = 0.97
Frameshift ()	TARGETRATE = 100000
Window Size	WINDOWSIZE = 250000
Using Hamming Window ()	USEHAMMING = T
No. of Filter Bank Channels ()	NUMCHANS = 26
No. of the Cepstral Coefficients	NUMCEPS = 12
Save the Output File Compressed	SAVECOMPRESSED = T