Dynamic Voice Parameter Modifications in Speech Signals

The aim of this paper is to present a system capable of modifying the human voice: volume by direct amplification/attenuation, duration of the voice signal and pitch using the Phase Vocoder, and timbre with the help of cepstral analysis. The system is also able to dynamically modify the aforementioned parameters in real-time. The proposed system was evaluated using a set of “clean” speech samples from the LibriSpeech ASR corpus of English speech with the Perceptual Evaluation of Speech Quality (PESQ) standard.