arXivDaily arXiv每日学术速递 周一至周五更新
看起来今天可以划水了
arXiv 2006.03575cs.SDcs.LGeess.AS
收藏

端到端对抗性文本到语音转换

End-to-End Adversarial Text-to-Speech

DeepMind(深度思维公司)

Jeff Donahue, Sander Dieleman, Mikołaj Bińkowski, Erich Elsen, Karen Simonyan

摘要:

现代文本到语音合成流水线通常涉及多个处理阶段,每个处理阶段都是独立于其余部分设计或学习的。在这项工作中,我们承担了一项具有挑战性的任务,即学习以端到端的方式从标准化文本或音素合成语音,从而产生直接对字符或音素输入序列进行操作并产生原始语音音频输出的模型。我们提出的生成器是前馈的,因此有效的训练和推理,使用基于令牌长度预测的可微分对齐方案。它通过对抗性反馈和预测损失的组合来学习产生高保真音频,从而限制生成的音频在其总持续时间和梅尔频谱图方面与地面实况大致匹配。为了允许模型捕获所生成的音频中的时间变化,我们在基于频谱的预测损失中采用软动态时间扭曲。由此产生的模型在5分制上实现了超过4的平均意见得分,这与依赖于多阶段训练和额外监督的最先进模型相当。

英文摘要:

Modern text-to-speech synthesis pipelines typically involve multiple processing stages, each of which is designed or learnt independently from the rest. In this work, we take on the challenging task of learning to synthesise speech from normalised text or phonemes in an end-to-end manner, resulting in models which operate directly on character or phoneme input sequences and produce raw speech audio outputs. Our proposed generator is feed-forward and thus efficient for both training and inference, using a differentiable alignment scheme based on token length prediction. It learns to produce high fidelity audio through a combination of adversarial feedback and prediction losses constraining the generated audio to roughly match the ground truth in terms of its total duration and mel-spectrogram. To allow the model to capture temporal variation in the generated audio, we employ soft dynamic time warping in the spectrogram-based prediction loss. The resulting model achieves a mean opinion score exceeding 4 on a 5 point scale, which is comparable to the state-of-the-art models relying on multi-stage training and additional supervision.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑