Multi-Reference Neural TTS Stylization with Adversarial Cycle Consistency

By M. Whitehill, Shuang Ma, Daniel J. McDuff, Yale Song. 2019

35 citations recorded by the source corpus.

Current multi-reference style transfer models for Text-to-Speech (TTS) perform sub-optimally on disjoints datasets, where one dataset contains only a single style class for one of the style dimensions. These models generally fail to produce style transfer for the dimension that is underrepresented in the dataset. In this paper, we propose an adversarial cycle consistency training scheme with paired and unpaired triplets to ensure the use of information from all style dimensions. During training, we incorporate unpaired triplets with randomly selected reference audio samples and encourage the synthesized speech to preserve the appropriate styles using adversarial cycle consistency. We use this method to transfer emotion from a dataset containing four emotions to a dataset with only a single emotion. This results in a 78% improvement in style transfer (based on emotion classification) with minimal reduction in fidelity and naturalness. In subjective evaluations our method was consistently rated as closer to the reference style than the baseline. Synthesized speech samples are available at: this https URL

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

1. Xu Tan
2. Tao Qin
3. Tie-Yan Liu
4. Frank K. Soong
5. Lei He
6. Xinyu Yang
7. Zhaoxi Mu
8. Yizhuo Dong

In Signal Processing

1. Xu Tan
2. Tao Qin
3. Tie-Yan Liu
4. Frank K. Soong
5. Lei He
6. Zhaoxi Mu
7. Xinyu Yang
8. Yizhuo Dong

Citations in the corpus

• A Survey on Neural Speech Synthesis (2021, methodology): “Technique Description Work Disentangling with Adversarial Training Disentanglement for control [218, 117, 275, 427] Cycle Consistency/Feedback for Control Enhance style/timbre generation [198, 380, 202, 30, 191] Semi-Supervised Learning for Control Use VAE and adversarial training [100, 116, 117, 42”
• Controllable Emotion Transfer For End-to-End Speech Synthesis (2020, unknown): “Following the principle of ‘say it like this’ [6], there has evolved a plenty of work in this direction lately [29], such as Global Style Tokens (GST) [7, 15], Variational Autoencoder (VAE) [8, 30] and their variants and updates [31, 32].”
• Improving Performance of Seen and Unseen Speech Style Transfer in End-to-end Neural TTS (2021, methodology): “• MRF-ACC: we adopt an adversarial cycle consistency training scheme for multi-reference neural TTS stylization [23] to ensure the use of information from all style classes and achieve style transfer.”
• Review of end-to-end speech synthesis technology based on deep learning (2021, methodology): “[229] used an adversarial cycle consistency training scheme to ensure the use of information from all style dimensions to address the challenges of multi-reference style transfer on disjoint datasets.”
• Cycle consistent network for end-to-end style transfer TTS training (2021, methodology): “rom all the above methods where it is always a totally paired eference speech used, there have been a few papers recently that se multiple reference inputs and/or introduce in an unpaired raining (Bian et al., 2019; Ma et al., 2019; Whitehill et al., 2020).”