Paris Smaragdis

Method of moments learning for left-to-right Hidden Markov models

2015

We propose a method-of-moments algorithm for parameter learning in Left-to-Right Hidden Markov Models. Compared to the conventional Expectation Maximization approach, the proposed algorithm is computationally more efficient, and hence more appropriate for large datasets. It …

Neural Speech Synthesis on a Shoestring: Improving the Efficiency of LPCNet

2022 · arXiv (Cornell University)

Neural speech synthesis models can synthesize high quality speech but typically require a high computational complexity to do so. In previous work, we introduced LPCNet, which uses linear prediction to significantly reduce the complexity of …