Adaptive processing of singing voice timbre

In this paper we propose an adaptive digital signal processing system that aims at achieving intuitive control over singing voice timbre. The sound transformations are operated using an analysis/synthesis system based on the harmonic plus noise model, from which singing voice timbre descriptors are extracted and related to common perceptual timbre attributes. Singing voice timbre modifications are achieved using the voice descriptors directly as transformation control parameters. An overview of the system is presented covering the techniques involved in the analysis, transformation and synthesis stages, along with some details on the software implementation. An example of a digital audio effect able to add a desired degree of breathiness to voice sounds is presented.