arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.08499eess.AScs.SD
收藏

基于生成对抗网络的多尺度优化近端语音可懂度增强

Multi-Metric Optimization using Generative Adversarial Networks for Near-End Speech Intelligibility Enhancement

National Institute of Informatics(国立情报学研究所);SOKENDAI(综合研究大学院大学)

Haoyu Li, Junichi Yamagishi

摘要:

在环境噪声和混响的存在下,语音的可懂度严重降低。在本文中,我们提出了一种新的基于深度学习的系统,用于修改语音信号以在等功率约束下提高其可懂度,即,修改前后的信号功率必须相同。为了实现这一点,我们使用生成对抗网络(GANs)来获得时频相关的放大因子,然后将其应用于输入的原始语音以重新分配语音能量。我们训练一个深度神经网络(DNN)模型,以同时优化多个高级语音指标,而不是只优化一个简单的指标,包括与清晰度和质量相关的指标,从而显著提高了性能和鲁棒性。我们的系统不仅可以工作在非实时模式下的离线音频播放,但也支持实际的实时语音应用。使用客观测量和主观听音测试的实验结果表明,该系统显着优于国家的最先进的基线系统在各种嘈杂和混响听音条件下。

英文摘要:

The intelligibility of speech severely degrades in the presence of environmental noise and reverberation. In this paper, we propose a novel deep learning based system for modifying the speech signal to increase its intelligibility under the equal-power constraint, i.e., signal power before and after modification must be the same. To achieve this, we use generative adversarial networks (GANs) to obtain time-frequency dependent amplification factors, which are then applied to the input raw speech to reallocate the speech energy. Instead of optimizing only a single, simple metric, we train a deep neural network (DNN) model to simultaneously optimize multiple advanced speech metrics, including both intelligibility- and quality-related ones, which results in notable improvements in performance and robustness. Our system can not only work in non-realtime mode for offline audio playback but also support practical real-time speech applications. Experimental results using both objective measurements and subjective listening tests indicate that the proposed system significantly outperforms state-ofthe-art baseline systems under various noisy and reverberant listening conditions.

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。

↑