Third International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications (MAVEBA 2003)

Florence, Italy
December 10-12, 2003

Sample-Based Singing Voice Synthesizer Using Spectral Models and Source-Filter Decomposition

J. Bonada (1), A. Loscos (1), O. Mayor (1), H. Kenmochi (2)

(1) Music Technology Group, Audiovisual Institute, Universitat Pompeu Fabra, Barcelona, Spain
(2) Advanced System Development Center, YAMAHA corporation, Hamamatsu, Japan

This paper is a review of the work contained in the insides of a sample-based virtual singing synthesizer. Starting with a narrative of the evolution of the techniques involved in it, the paper focuses mainly on the description of its current components and processes and its most relevant features: from the singer databases creation to the final synthesis concatenation step.

Full Paper (reprinted with permission from Firenze University Press)

Bibliographic reference.  Bonada, J. / Loscos, A. / Mayor, O. / Kenmochi, H. (2003): "Sample-based singing voice synthesizer using spectral models and source-filter decomposition", In MAVEBA-2003, 175-178.