Pretraining Strategies, Waveform Model Choice, and Acoustic Configurations for Multi-Speaker End-to-End Speech Synthesis

Autor:	Cooper, Erica, Wang, Xin, Zhao, Yi, Yasuda, Yusuke, Yamagishi, Junichi
Rok vydání:	2020
Předmět:	Computer Science - Sound Computer Science - Computation and Language
Druh dokumentu:	Working Paper
Popis:	We explore pretraining strategies including choice of base corpus with the aim of choosing the best strategy for zero-shot multi-speaker end-to-end synthesis. We also examine choice of neural vocoder for waveform synthesis, as well as acoustic configurations used for mel spectrograms and final audio output. We find that fine-tuning a multi-speaker model from found audiobook data that has passed a simple quality threshold can improve naturalness and similarity to unseen target speakers of synthetic speech. Additionally, we find that listeners can discern between a 16kHz and 24kHz sampling rate, and that WaveRNN produces output waveforms of a comparable quality to WaveNet, with a faster inference time. Comment: Technical report
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2011.04839 Zobrazit plný text záznamu View this record from Arxiv