基于DNN的蒙特卡罗丢弃的语音增强
DNN Based Speech Enhancement for Unseen Noises Using Monte Carlo Dropout
摘要：
在这项工作中，我们提出使用辍学作为贝叶斯估计器，以提高深度神经网络（DNN）的泛化能力，用于语音增强。通过使用蒙特卡罗（MC）丢弃，我们表明DNN在不可见噪声和SNR条件下具有更好的增强效果。DNN在SNR为0，5和10 dB的Factory 2，M109，Babble，Leopard和Volvo噪声中进行语音训练，并在白色，粉红色和factory 1噪声中进行语音测试。语音样本来自TIMIT数据库，噪声来自NOISEX-92。在另一个实验中，我们训练五个DNN模型，分别对Factory 2，M109，Babble，Leopard和Volvo噪声，在0，5和10 dB SNR下的语音进行破坏。模型精度（使用MC dropout估计）用作平方误差的代理，以根据DNN模型在每帧测试数据上的性能动态选择最佳DNN模型。
英文摘要：
In this work, we propose the use of dropouts as a Bayesian estimator for increasing the generalizability of a deep neural network (DNN) for speech enhancement. By using Monte Carlo (MC) dropout, we show that the DNN performs better enhancement in unseen noise and SNR conditions. The DNN is trained on speech corrupted with Factory2, M109, Babble, Leopard and Volvo noises at SNRs of 0, 5 and 10 dB and tested on speech with white, pink and factory1 noises. Speech samples are obtained from the TIMIT database and noises from NOISEX-92. In another experiment, we train five DNN models separately on speech corrupted with Factory2, M109, Babble, Leopard and Volvo noises, at 0, 5 and 10 dB SNRs. The model precision (estimated using MC dropout) is used as a proxy for squared error to dynamically select the best of the DNN models based on their performance on each frame of test data.
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。
