Bidirectional Variational Inference for Non-Autoregressive Text-to-Speech

By Yoonhyung Lee, Joongbo Shin, Kyomin Jung. 2021

42 citations recorded by the source corpus.

Subfields: Natural Language Generation, Speech & Audio

How researchers see this paper

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

In Natural Language Generation

1. Dong Yu
2. Dan Su
3. Ye Jia
4. Frank K. Soong
5. Jungil Kong
6. Xie Chen
7. Sungwon Kim
8. Heeseung Kim

In Speech & Audio

1. Xu Tan
2. Tao Qin
3. Yi Ren
4. Zhou Zhao
5. Tie-Yan Liu
6. Jinglin Liu
7. Rongjie Huang
8. Sheng Zhao

Citations in the corpus

• ELLA-V: Stable Neural Codec Language Modeling with Alignment-guided Sequence Reordering (2024, methodology): “Some methods, utilizing techniques such as VAE (Hsu et al., 2019; Lee et al., 2022), flow (Miao et al., 2020; Kim et al., 2020), diffusion (Jeong et al., 2021; Kim et al., 2022; Popov et al., 2021), and others (Wu and Shi, 2022), have achieved promising performance in end-to-end speech synthesis.”
• Mega-TTS: Zero-Shot Text-to-Speech at Scale with Intrinsic Inductive Bias (2023, methodology): “Generative models, like language models [4, 33], VAE [34, 47], GAN [31, 30], Normalizing flow [39, 29], and diffusion model [32, 24, 43, 22], have been applied to speech or audio synthesis for years.”
• Guided-TTS: A Diffusion Model for Text-to-Speech via Classifier Guidance (2021, methodology): “In addition, variational autoencoder based models (Kingma & Welling, 2014; Lee et al., 2020; Liu et al., 2021), diffusion based models (Ho et al.”
• A Survey on Neural Speech Synthesis (2021, methodology): “, Flow, GAN, VAE, Diffusion) Besides the above acoustic models, there are a lot of other acoustic models [361, 22, 123, 183, 54], as shown in Table 2.”
• Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech (2021, methodology): “Parallel TTS models (Ren et al., 2019; Peng et al., 2020; Kim et al., 2020; Ren et al., 2021; Lee et al., 2021), on the other hand, have been relied on deterministic duration prediction.”
• Diff-TTS: A Denoising Diffusion Model for Text-to-Speech (2021, methodology): “The dilations of the convolution are [1, 2, 4, 1, 2, 4, 1, 2, 4, 1] from the bottom to the top, and the kernel size is 4 with ReLU activation function followed by layer normalization [26].”
• DiffGAN-TTS: High-Fidelity and Efficient Text-to-Speech with Denoising Diffusion GANs (2022, methodology): “Other generative models have also been studied for TTS, such as Flow-based models(Kim et al., 2020; Miao et al., 2020), variational auto-encoder (VAE)-based models(Lee et al., 2021; Liu et al., 2021b), and generative adversarial network (GAN)-based models (Donahue et al., 2021; Yang et al., 2021).”
• StyleTTS: A Style-Based Generative Model for Natural and Diverse Text-to-Speech Synthesis (2022, unknown): “While several strategies have been proposed to address this, including methods based on variational inference [1], [5], [6], [7], flow-based modeling [1], [8], [9], explicit modeling of pitch, duration and energy [10], [11], [12], autoregressive language modeling (LM) objectives [13], [14], [15],…”
• Translatotron 2: High-quality direct speech-to-speech translation with voice preservation (2021, methodology): “While most of these works conducted experiments using synthetic datasets with translation speech in a clean single speaker’s voice, Jia et al. (2019b); Lee et al. (2021a) conducted experiments using multi-speaker human recordings.”
• Translatotron 2: Robust direct speech-to-speech translation (2021, methodology): “C N ON - AUTOREGRESSIVE SYNTHESIZER Following recent non-autoregressive TTS works (Ren et al., 2021; Guo et al., 2021; Lee et al., 2021; Elias et al., 2021a), we explored using non-autoregressive synthesizer in Translatotron 2, which may have significantly lower latency at inference time.”
• VARA-TTS: Non-Autoregressive Text-to-Speech Synthesis based on Very Deep VAE with Residual Attention (2021, methodology): “…module needs duration labels as supervision, which may come from pre-trained AR TTS models (Ren et al., 2019; Peng et al., 2020), forced aligner (Ren et al., 2020), dynamic programming backtrack path (Kim et al., 2020) or jointly-trained attention modules (Miao et al., 2020; Lee et al., 2021).”