Wavelets proves to be an effective method in analyzing speech signals that contain both steady state characteristics (vowels) and transient characteristics (consonants), since different combinations of vowels and consonants have distinct characteristics in different octaves.

If more time were available, we are sure that we can find even more differences in different octaves between each of the digits, therefore giving even better results. Better results can probably be obtained also combining our project with Digital Bubblebath -- the 1996 speech recognition project, which would add formants and pitch determinations to the wavelet analysis.

Possible future ideas to continue this project:

• Experiment with the continuous wavelet transform which steps though the frequencies and times continuously (rmdwt.mexsol in Rice Matlab Toolbox), instead of the discreate wavelet transform used in this project.

• Find a method to break up words like "twenty-one" which will make possible the recognition of all intergers.

• Integrate this project with Digital Bubblebath.

• Extending the digits to 20.