Ganspeech: Adversarial Training For High-fidelity Multi-speaker Speech Synthesis

Abstract

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. Fine-tuning to the target speaker data with the multi-speaker model can achieve better quality, however, there still exists a gap compared to the real speech sample and the model depends on the speaker. In this work, we propose GANSpeech, which is a high-fidelity multi-speaker TTS model that adopts the adversarial training method to a non-autoregressive multi-speaker TTS model. In addition, we propose simple but efficient automatic scaling methods for feature matching loss used in adversarial training. In the subjective listening tests, GANSpeech significantly outperformed the baseline multi-speaker FastSpeech and FastSpeech2 models, and showed a better MOS score than the speaker-specific fine-tuned FastSpeech2.

Stats

• citations21
• S2 citations—
• github stars0
• HF likes0
• heat score10.07
• arxiv keyyang2021ganspeech

Related papers

• Multi-task Adversarial Training Algorithm For Multi-speaker Neural Text-to-speech (2022)0.00
• High Fidelity Speech Synthesis With Adversarial Networks (2019)0.00
• Multi-spectrogan: High-diversity And High-fidelity Spectrogram Generation With Adversarial Style Combination For Speech Synthesis (2020)0.00
• Statistical Parametric Speech Synthesis Using Generative Adversarial Networks Under A Multi-task Learning Framework (2017)10.21
• Statistical Parametric Speech Synthesis Incorporating Generative Adversarial Networks (2017)16.21
• Hifi-gan: Generative Adversarial Networks For Efficient And High Fidelity Speech Synthesis (2020)0.00
• Waveform Generation For Text-to-speech Synthesis Using Pitch-synchronous Multi-scale Generative Adversarial Networks (2018)8.35
• Expediting TTS Synthesis With Adversarial Vocoding (2019)6.77