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 1803.09016eess.AScs.SD
收藏

一种改进的基于DNN的谱特征映射,用于鲁棒的自动语音识别

An improved DNN-based spectral feature mapping that removes noise and reverberation for robust automatic speech recognition

Universidad de Chile(智利大学);Universidad de la Frontera(拉弗龙特拉大学);Carnegie Mellon University(卡内基梅隆大学)

Juan Pablo Escudero, José Novoa, Rodrigo Mahu, Jorge Wuth, Fernando Huenupán, Richard Stern, Néstor Becerra Yoma

摘要:

混响和加性噪声对自动语音识别系统的性能有不利影响。在本文中,我们探讨了基于DNN的频谱特征映射的能力,以消除混响和加性噪声的影响。使用CHiME-2数据库的实验表明,与基线系统相比,该DNN在SNR等于-6 dB,-3 dB,0 dB和3 dB时可以实现WER平均降低4.5%,并且在SNR大于6 dB和9 dB时仅降低0.8%。这些结果表明,该DNN在去除加性噪声方面比混响更有效。为了提高DNN的性能,我们将其与加权预测误差(WPE)方法相结合,该方法显示出互补的行为。虽然与基线相比,这种组合使WER降低了约11%,但观察到的改善不如单独使用WPE获得的改善那么大。然而,对DNN训练过程进行了修改,与基线系统相比,WER平均降低了18.3%。此外,与单独的WPE相比,改进的DNN结合WPE实现了7.9%的WER降低。

英文摘要:

Reverberation and additive noise have detrimental effects on the performance of automatic speech recognition systems. In this paper we explore the ability of a DNN-based spectral feature mapping to remove the effects of reverberation and additive noise. Experiments with the CHiME-2 database show that this DNN can achieve an average reduction in WER of 4.5%, when compared to the baseline system, at SNRs equal to -6 dB, -3 dB, 0 dB and 3 dB, and just 0.8% at greater SNRs of 6 dB and 9 dB. These results suggest that this DNN is more effective in removing additive noise than reverberation. To improve the DNN performance, we combine it with the weighted prediction error (WPE) method that shows a complementary behavior. While this combination provided a reduction in WER of approximately 11% when compared with the baseline, the observed improvement is not as great as that obtained using WPE alone. However, modifications to the DNN training process were applied and an average reduction in WER equal to 18.3% was achieved when compared with the baseline system. Furthermore, the improved DNN combined with WPE achieves a reduction in WER of 7.9% when compared with WPE alone.

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。

↑