Evaluation of noisy speech recognition based on noise reduction and acoustic model adaptation on the Aurora2 tasks
Evaluation of noisy speech recognition based on noise reduction and acoustic model adaptation on the Aurora2 tasks
复制标题
基于 Aurora2 任务的降噪和声学模型自适应的噪声语音识别评估
DOI：
10.21437/icslp.2002-20
复制
发表时间：
2002
期刊：
影响因子：
--
通讯作者：
Y. Ariki
中科院分区：
文献类型：
--
作者：
Masakiyo Fujimoto;Y. Ariki
在本文中，我们在 AURORA2 任务上评估了基于降噪和声学模型自适应的噪声语音识别方法。对于降噪方法，我们采用了两种降噪方法。一种是自适应子带频谱减法（ASBSS）方法，该方法可以根据每帧频带中的SNR来优化噪声减除率。另一种是卡尔曼滤波估计方法，它根据 ABSS 估计的语音频谱重新估计准确的语音频谱。通过结合这些方法来估计准确的语音频谱。通常，降噪方法存在这样的问题：由于通过降噪和过估计而产生的残余噪声导致谱失真，因此降低了识别率。为了解决降噪方法中的问题，通过使用对频谱失真的无监督 MLLR 自适应来实现声学模型的自适应。在对 AURORA2 任务的评估中，我们的方法显示在干净训练条件和多训练条件下识别精度都有显着提高。
In this paper, we have evaluated a noisy speech recognition method based on noise reduction and acoustic model adaptation, on the AURORA2 tasks. For noise reduction method, we employed two noise reduction methods. One is an Adaptive Sub-Band Spectral Subtraction (ASBSS) method which can optimize the noise subtraction rate according to the SNR in frequency bands at each frame. The other is a Kalman filtering estimation method which re-estimates the accurate speech spectra from those estimated by ASBSS. The accurate speech spectra was estimated by combining these methods. Usually, a noise reduction method has a problem that it degrades the recognition rate because of spectral distortion caused by residual noise occurred through noise reduction and over estimation. To solve the problem in noise reduction method, adaptation of the acoustic models is employed by using an unsupervised MLLR adaptation to the spectral distortion. In evaluation on the AURORA2 tasks, our method showed the significant improvement in recognition accuracy for both clean training condition and multi training condition.
影响因子：
4.3
作者：
LEGGETTER, CJ;WOODLAND, PC
通讯作者：
WOODLAND, PC