View Full Version : Speech Analysing algorithm

March 11th, 2008, 12:35 PM
I want to analyse and compare a single spoken word e.g. 'apple' with a template of the acceptable (correct) sound for that word.
I need to be able to vary the parameters for what is 'acceptable' so that initially any sound may be accepted and appropriate 'reward' given.
I then want to gradually decrease the tolerance given so 'rewards' will only be given as the subjects response more closely approximates the template.

I am aware of PRAAT system (Fantastic for analysing sounds) but I want to be able to incorporate this into a training program and need the algorithm within the program.