PROJECT TITLE :
Deep Learning for Acoustic Modeling in Parametric Speech Generation: A systematic review of existing techniques and future trends
Hidden Markov models (HMMs) and Gaussian mixture models (GMMs) are the 2 commonest types of acoustic models employed in statistical parametric approaches for generating low-level speech waveforms from high-level symbolic inputs via intermediate acoustic feature sequences. However, these models have their limitations in representing advanced, nonlinear relationships between the speech generation inputs and therefore the acoustic features. Galvanized by the intrinsically hierarchical process of human speech production and by the successful application of deep neural networks (DNNs) to automatic speech recognition (ASR), deep learning techniques have also been applied successfully to speech generation, as reported in recent literature. This text systematically reviews these rising speech generation approaches, with the twin goal of helping readers gain a higher understanding of the prevailing techniques in addition to stimulating new work within the burgeoning space of deep learning for parametric speech generation.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here