no code implementations • 27 Feb 2023 • Yoonhyung Lee, Jinhyeok Yang, Kyomin Jung
Also, the objective function of NF makes the model use the variance information and the text in a disentangled manner resulting in more precise variance control.
2 code implementations • 27 Jun 2022 • Taejun Bak, Junmo Lee, Hanbin Bae, Jinhyeok Yang, Jae-Sung Bae, Young-Sun Joo
Therefore, in this paper, we investigate the relationship between these artifacts and GAN-based vocoders and propose a GAN-based vocoder, called Avocodo, that allows the synthesis of high-fidelity speech with reduced artifacts.
no code implementations • 8 Apr 2022 • Jae-Sung Bae, Jinhyeok Yang, Tae-Jun Bak, Young-Sun Joo
This paper proposes a hierarchical and multi-scale variational autoencoder-based non-autoregressive text-to-speech model (HiMuV-TTS) to generate natural speech with diverse speaking styles.
no code implementations • 29 Jun 2021 • Jinhyeok Yang, Jae-Sung Bae, Taejun Bak, Youngik Kim, Hoon-Young Cho
Recent advances in neural multi-speaker text-to-speech (TTS) models have enabled the generation of reasonably good speech quality with a single model and made it possible to synthesize the speech of a speaker with limited training data.
2 code implementations • 30 Jul 2020 • Jinhyeok Yang, Jun-Mo Lee, Youngik Kim, Hoon-Young Cho, Injung Kim
Additionally, compared with Parallel WaveGAN, another recently developed high-fidelity vocoder, VocGAN is 6. 98x faster on a CPU and exhibits higher MOS.