arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1503.02108cs.LGcs.CLcs.NE
收藏

深度模型中网络参数的最大后验自适应

Maximum a Posteriori Adaptation of Network Parameters in Deep Models

Georgia Institute of Technology(佐治亚理工学院);Kore University of Enna(恩纳科雷大学);Microsoft Corporation(微软公司)

Zhen Huang, Sabato Marco Siniscalchi, I-Fan Chen, Jiadong Wu, Chin-Hui Lee

摘要:

我们提出了一种贝叶斯方法来调整训练有素的上下文相关的深度神经网络隐马尔可夫模型(CD-DNN-HMM)的参数,以提高自动语音识别性能。给定丰富的DNN参数,但只有有限的数据量,适应DNN模型的有效性往往会受到影响。我们制定了专门设计的CD-DNN-HMM参数的最大后验(MAP)自适应,其中增强的线性隐藏网络连接到输出绑定状态或senone,并将其与之前提出的特征空间MAP线性回归进行比较。在20,000词开放词汇量的《华尔街日报》任务上的实验证据证明了该框架的可行性。在监督自适应中,所提出的MAP自适应方法提供了超过10%的相对误差减少,并始终优于传统的基于变换的方法。此外,我们提出了一个初步的尝试,通过利用senone之间的相似性,产生层次先验,以提高适应效率和有效性与有限的适应数据。

英文摘要:

We present a Bayesian approach to adapting parameters of a well-trained context-dependent, deep-neural-network, hidden Markov model (CD-DNN-HMM) to improve automatic speech recognition performance. Given an abundance of DNN parameters but with only a limited amount of data, the effectiveness of the adapted DNN model can often be compromised. We formulate maximum a posteriori (MAP) adaptation of parameters of a specially designed CD-DNN-HMM with an augmented linear hidden networks connected to the output tied states, or senones, and compare it to feature space MAP linear regression previously proposed. Experimental evidences on the 20,000-word open vocabulary Wall Street Journal task demonstrate the feasibility of the proposed framework. In supervised adaptation, the proposed MAP adaptation approach provides more than 10% relative error reduction and consistently outperforms the conventional transformation based methods. Furthermore, we present an initial attempt to generate hierarchical priors to improve adaptation efficiency and effectiveness with limited adaptation data by exploiting similarities among senones.

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。

↑