The purpose of the project is to show how a discrete time system model can be related to a speech waveform. Please refer 'System-model.jpg' to look at the system model. In this project, I used the exponential and the Rosenberg models to represent the glottal waveform. The vocal tract is modeled as concatenated lossless tube. The final goal is to generate periodic vowel sounds such as 'AA' and 'IY'. This project was done in Matlab.
Sundarram/Speech-Signal-Processing
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|