All you Terminator fans out there would be quite familiar with voice transformation technology. You would have noticed the terminator imitating someone’s voice perfectly.
Leave aside Sci-Fi, the U.S. Air Force is keen to get such a technology in reality. Technologies that can help its pilots change their voices and sound like another person altogether.
Such a technology is not that difficult to build as it sounds and can be accomplished with some sound transformation algorithms.
The original requirements from the U.S.A.F. solicitation are:
The goal of this phase is to research techniques to analyze a person [sic] voice for voice transformation. While voice transformation have [sic] been around for awhile, the ability [sic] to transform a person’s voice to a target voice is not yet solved. Parameters such as the speaking rate, stress, and intonation will provide broad parameters for modeling a person’s voice. A finer grain analysis of a person’s voice may also be performed by de-convolving an audio signal into its glottal pulse and vocal tract information.
Transforming a speaker’s voice to make it unidentifiable is less difficult as the process of transforming someone’s voice into another person’s voice. The main problem areas in such a technology include:
• Formant spectra:
Formant refers to the regions of concentrating of energy, prominent on a sound spectrogram. This collectively constitute the frequency spectrum of a speech sound.
• Prosodic Features:
These are those aspects of speech that vary from person to person, like fundamental pitch of voice, timing and the different patterns of speech.
• Mannerisms:
This refers to the word choices and the use of phrases that a person uses while talking.
The U.S.A.F. is looking for a variety of uses for such a technology. For example a person with a damaged voice box can surely benefit from this technology and this can also be useful in the gaming industry.
Via: technovelgy
