Est. Budget: $4,000.00
I am looking to have a text to speech system developed which is based on a mathematical model of the vocal tract. The speech should be entirely synthetic (no prerecorded voice clips), which gives a lot of control over vocal characteristics such as head size, speaking rate, inflection/intonation, breathyness, and different genders. The system should use the source filter model where speech is generated from an excitation signal that is made up of white noise for the unvoiced parts ...