When creating a steady tone sound, humans close their vocal cords to varying degrees. As the vocal cords are held progressively farther apart, more air leak occurs during tone production, creating more white noise. I would like to be able to extract the ratio of pure tone to white noise from an audio recording.

I would appreciate input from an audio engineer about how this can be mathematically accomplished. Is there already a framework, such as in Matlab, to obtain such a computation?