KaleiVoiceCope is the name of a Real-time Voice Transformation Technology developed by the Music Technology Group of The Pompeu Fabra University. An input voice is first analyzed in the spectral domain and some spectral descriptors are then extracted. Based on a set of parameters, a new voice is generated changing its timbre, amplitude, pitch, and other spectral and physical characteristics. This transformation allows a wide range of possibilities, for instance, changing the gender of a voice from male to female or transforming a teenager to an old woman but also more exotic transformations are possible such as robotizing the voice, converting the voice in order to be used in a cartoon character or giving the voice an alien character as it was taken from a science fiction film.
The word "KaleiVoiceCope" comes from combining the words Kaleidoscope and Voice. A kaleidoscope is a tube-shaped optical instrument that is rotated to produce a succession of symmetrical designs by means of mirrors reflecting the constantly changing patterns made by bits of colored glass at one end of the tube. The idea here is to replace the bits of colored glass by the human voice, so a KaleiVoiceCope is used to transform the voice character, tone, pitch, gender, timbre, etc based on a set of controllable parameters.
The KaleiVoiceCope Technology consists in a Software Library containing state of the art signal processing algorithms that convert and modify the human voice, based in a set of meaningful controls that preserve its natural quality.
These meaningful transformations include adding vibrato, changing fundamental frequency or amplitude, controlling spectral and physical characteristics of the voice or modifying timbre. High-Level Presets based on these transformations can be easily created to allow, for instance, gender change transformations, convert to operatic voice or robotizer effect.
This Voice Transformation Library can be easily integrated in post-production studios or software FX plug-ins to transform isolated singing or speech voices from a song, film or advertisement or even generate new virtual characters from existing ones.
This technology can be licensed through our spin-off company Voctro Labs.