arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1806.00511eess.AScs.SDeess.SP
收藏

基于性能的端到端语音分离代价函数

Performance Based Cost Functions for End-to-End Speech Separation

University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Adobe Research(奥多比研究院)

Shrikant Venkataramani, Ryley Higa, Paris Smaragdis

摘要:

最近用于源分离的神经网络策略试图通过直接处理音频信号的波形来对音频信号进行建模。均方误差(MSE)测量去噪语音的波形与地面实况语音的波形之间的欧几里得距离,其已经是这些方法的自然成本函数。然而,MSE不是感知动机的测量,并且可能导致大的感知差异。在本文中,我们提出并实验了新的损失函数的端到端的源分离。这些损失函数由BSS评估和感知度量(如源失真比(SDR)、源干扰比(SIR)、源伪影比(SAR)和短时客观可懂度比(STOI))激发。这使得能够根据任务的要求灵活地混合和匹配这些损失函数。主观听音测试表明,建议的代价函数的组合有助于实现优越的分离性能相比,独立的MSE和SDR成本。

英文摘要:

Recent neural network strategies for source separation attempt to model audio signals by processing their waveforms directly. Mean squared error (MSE) that measures the Euclidean distance between waveforms of denoised speech and the ground-truth speech, has been a natural cost-function for these approaches. However, MSE is not a perceptually motivated measure and may result in large perceptual discrepancies. In this paper, we propose and experiment with new loss functions for end-to-end source separation. These loss functions are motivated by BSS\_Eval and perceptual metrics like source to distortion ratio (SDR), source to interference ratio (SIR), source to artifact ratio (SAR) and short-time objective intelligibility ratio (STOI). This enables the flexibility to mix and match these loss functions depending upon the requirements of the task. Subjective listening tests reveal that combinations of the proposed cost functions help achieve superior separation performance as compared to stand-alone MSE and SDR costs.

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。

↑