Summary: This application will take an incoming voice stream and using a “code file” alter and stream the voice out in near real-time. The voice changing application will "call" the code file on demand so it may convert the incoming stream to any target voice for which there is a code file.
The code file will be created by the application (or a standalone application) in “training mode” having a person speaking the “target voice” reading and pronouncing a repertoire of words and phrases.
The application will run on Linux. Multiple instances of the application may be run on the machine at the same time. I envision that one would call the application from the command line indicating the input and output stream "device" and the code-file to use. Other means of initiating the application will be considered. It is understood that the application can not change the words being spoken, it will only change the sound of the words and sound of the voice. It may change pauses, cadence, pitch, timbre, and other voice characteristics to mimic the target voice.
The algorithm and approach to the problem is not specified. The developer may use existing libraries or create his/her own. The objective is to be able to speak into the application and have the words come out of the application in the target voice. This project includes creating the voice changing application and the "training" application that creates the code file. The programming language to be used in not specified.
Please respond with the approach you would take to accomplish the project. What format(s) the input and output streams would need to take (if it is limited), what experience you have in this area of development, and what you think the limitations on the output will be with your solution.
5 freelancers are bidding on average $1305 for this job
I can do your work in Qt frameworking in C++. I got 9 years experience in software development. Please go thru my profile and let me know if interested. Thanks Sujan