Dictionary Update for NMF-based Voice Conversion Using an Encoder-Decoder Network
This is an incremental improvement for voice conversion applications like personalized speech synthesis, addressing a specific bottleneck in dictionary efficiency.
The paper tackles the problem of needing large dictionaries for high-quality voice conversion in exemplar-based Nonnegative Matrix Factorization (NMF) by proposing a dictionary update method using an encoder-decoder network, resulting in significant gains with small dictionaries over conventional systems.
In this paper, we propose a dictionary update method for Nonnegative Matrix Factorization (NMF) with high dimensional data in a spectral conversion (SC) task. Voice conversion has been widely studied due to its potential applications such as personalized speech synthesis and speech enhancement. Exemplar-based NMF (ENMF) emerges as an effective and probably the simplest choice among all techniques for SC, as long as a source-target parallel speech corpus is given. ENMF-based SC systems usually need a large amount of bases (exemplars) to ensure the quality of the converted speech. However, a small and effective dictionary is desirable but hard to obtain via dictionary update, in particular when high-dimensional features such as STRAIGHT spectra are used. Therefore, we propose a dictionary update framework for NMF by means of an encoder-decoder reformulation. Regarding NMF as an encoder-decoder network makes it possible to exploit the whole parallel corpus more effectively and efficiently when applied to SC. Our experiments demonstrate significant gains of the proposed system with small dictionaries over conventional ENMF-based systems with dictionaries of same or much larger size.