Gaussian Lpcnet for Multisample Speech Synthesis

By Vadim Popov, Mikhail Kudinov, Tasnima Sadekova. 2020

14 citations recorded by the source corpus.

LPCNet vocoder has recently been presented to TTS community and is now gaining increasing popularity due to its effectiveness and high quality of the speech synthesized with it. In this work, we present a modification of LPCNet that is 1.5x faster, has twice less non-zero parameters and synthesizes speech of the same quality. Such enhancement is possible mostly due to two features that we introduce into the original architecture: the proposed vocoder is designed to generate 16-bit signal instead of 8-bit µ-companded signal, and it predicts two consecutive excitation values at a time independently of each other. To show that these modifications do not lead to quality degradation we train models for five different languages and perform extensive human evaluation.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Signal Processing

Citations in the corpus

A Survey on Neural Speech Synthesis (2021, background): “Some following works further improve LPCNet from different perspectives, such as reducing complexity for speedup [364, 269, 148], and improving stability for better quality [126].”
Lightweight LPCNet-Based Neural Vocoder with Tensor Decomposition (2020, methodology): “Furthermore, we can also combine our approach with prior works such as multisample generation [13].”
Fast and Lightweight On-Device TTS with Tacotron2 and LPCNet (2020, background): “While several solutions referenced above [7, 8, 9, 10] were claimed to be capable of faster-than-realtime on-device speech generation the requirements to the hardware are still high and can be met only on higher-end devices which are still too expensive for many people.”
Bunched LPCNet : Vocoder for Low-cost Neural Text-To-Speech Systems (2020, background): “In recently proposed Gaussian LPCNet [15], two samples are generated per inference by assuming independence for these samples.”