Voice Conversion Based On Hidden Markov Model

Voice morphing which is also referred to as voice transformation and voice conversion is a technique to modify a source speaker's speech utterance to sound as if it was spoken by a target speaker. There are many applications which may benefit from this sort of technology. In the paper, we propose a new method of voice conversion which uses Hidden Markov Model (HMM) for the training. HMM is used to represent the phonetic structure of training speech and to generate the training pairs of source and target speakers by mapping the HMM states between source and target speeches. Then, HMM codebook is generated to create the mapping function for the voice conversion.