Going back to the source : inverse filtering of the speech signal with ANNs
Joachim Denzler, Ralf Kompe, Andreas Kießling, Heinrich Niemann, Elmar Nöth · 1994
In this paper we present a new method transforming speech signals to voice source signals (VSS) using arti-cial neural networks (ANN). We will point out that the ANN mapping of speech signals into source signals is quite accurate, and most of the irregularities in the speech sig-nal will lead to an irregularity in the source signal, pro-duced by the ANN (ANN-VSS). We will show that the mapping of the ANN is robust with respect to untrained speakers, dierent recording conditions and facilities, and dierent vocabularies. We will also present preliminary results which show that from the ANN source signal pitch periods can be determined accurately.