Neural TTS Stylization with Adversarial and Collaborative Games
By Shuang Ma, Daniel J. McDuff, Yale Song. 2018
31 citations recorded by the source corpus.
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
Citations in the corpus
- A Survey on Neural Speech Synthesis (2021, methodology): “In this subsection, we review techniques on disentangling [218, 117, 275], controlling [353, 180, 240, 13, 267, 343, 192], and transferring [150, 131, 393, 6] variation information, as shown in Table 15.”
- Self-supervised Context-aware Style Representation for Expressive Speech Synthesis (2022, unknown): “Many recent studies in this area focus on these two issues [6, 7, 8].”
- Review of end-to-end speech synthesis technology based on deep learning (2021, methodology): “Moreover, they adopted a style loss [54, 88] to measure the style differences between the generated and reference mel-spectrogram [55, 134].”
- Unpaired Image-to-Speech Synthesis With Multimodal Information Bottleneck (2019, background): “Recent advances in deep generative models have shown impressive results across many cross-modal generation tasks, including text-to-image [36], text-to-speech [27], image-to-video [29], video-to-sound [56] synthesis.”
- Cycle consistent network for end-to-end style transfer TTS training (2021, methodology): “However, it is a shared style encoder in Ma et al. (2019). • The style transfer in the proposed method is in an unsupervised manner where the style label is not needed.”
- Multi-Reference Neural TTS Stylization with Adversarial Cycle Consistency (2019, methodology): “[3] and train for an additional 1k epochs.”
- Controllable Emotion Transfer For End-to-End Speech Synthesis (2020, unknown): “Recently, the Gram matrix has been adopted to measure mel-spectrogram of audio signals [31], aiming to capture local statistics of an audio signal in the frequency-time domain.”