arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1912.02461cs.SDcs.LGeess.AS
收藏

用于语音生成的鲁棒神经语音编码研究综述

Towards Robust Neural Vocoding for Speech Generation: A Survey

College of Electrical Engineering and Computer Science, National Taiwan University(台湾大学电机资讯学院);Graduate Institute of Communication Engineering, National Taiwan University(台湾大学电信工程学研究所)

Po-chun Hsu, Chun-hsuan Wang, Andy T. Liu, Hung-yi Lee

摘要:

近年来,神经声码器已广泛应用于语音合成任务,包括文本到语音和语音转换。然而,当遇到训练和推理之间的数据分布不匹配时,在真实数据上训练的神经声码器通常会在不可见的场景中降低语音质量。本文在五个不同的数据集上交替训练了WaveNet、WaveRNN、FFTNet、Parallel WaveGAN四种常用的神经元声码器。为了研究神经声码器的鲁棒性,我们使用来自可见/不可见扬声器,可见/不可见语言,文本到语音模型和语音转换模型的声学特征来评估模型。我们发现,对于实现通用声码器来说,说话者的多样性比语言更重要。通过我们的实验,我们表明WaveNet和WaveRNN更适合于文本到语音模型,而并行WaveGAN更适合于语音转换应用。大量的主观MOS结果的自然度的所有声码器,为未来的研究。

英文摘要:

Recently, neural vocoders have been widely used in speech synthesis tasks, including text-to-speech and voice conversion. However, when encountering data distribution mismatch between training and inference, neural vocoders trained on real data often degrade in voice quality for unseen scenarios. In this paper, we train four common neural vocoders, including WaveNet, WaveRNN, FFTNet, Parallel WaveGAN alternately on five different datasets. To study the robustness of neural vocoders, we evaluate the models using acoustic features from seen/unseen speakers, seen/unseen languages, a text-to-speech model, and a voice conversion model. We found out that the speaker variety is much more important for achieving a universal vocoder than the language. Through our experiments, we show that WaveNet and WaveRNN are more suitable for text-to-speech models, while Parallel WaveGAN is more suitable for voice conversion applications. Great amount of subjective MOS results in naturalness for all vocoders are presented for future studies.

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。

↑