arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1509.02412cs.CL
收藏

语音识别中基于潜在Dirichlet分配的无监督领域发现

Unsupervised Domain Discovery using Latent Dirichlet Allocation for Acoustic Modelling in Speech Recognition

University of Sheffield(谢菲尔德大学)

Mortaza Doulaty, Oscar Saz, Thomas Hain

摘要:

语音识别系统通常是高度依赖于域的,这是文献中广泛报道的事实。然而,域的概念是复杂的,不受明确标准的约束。因此,通常不清楚数据是否应被视为域外数据。虽然声学模型和语言模型都可以是特定于领域的,但本文的工作集中在声学建模上。我们提出了一种新的方法来执行无监督发现域使用潜在狄利克雷分配(LDA)建模。这里,假设数据中存在一组隐藏域,由此每个音频片段可以被认为是域属性的加权混合。将音频段分类到域中允许创建用于自动语音识别的域特定声学模型。实验是在一个包含广播电视、电话交谈、会议、讲座和朗读等多种语音数据的数据集上进行的,联合训练集为60小时,测试集为6小时。最大后验概率(MAP)适应LDA为基础的领域,以产生相对字出错率(WER)的改善高达16%的相对,相比,汇集的培训,和高达10%,与人类标记的领域知识适应的模型。

英文摘要:

Speech recognition systems are often highly domain dependent, a fact widely reported in the literature. However the concept of domain is complex and not bound to clear criteria. Hence it is often not evident if data should be considered to be out-of-domain. While both acoustic and language models can be domain specific, work in this paper concentrates on acoustic modelling. We present a novel method to perform unsupervised discovery of domains using Latent Dirichlet Allocation (LDA) modelling. Here a set of hidden domains is assumed to exist in the data, whereby each audio segment can be considered to be a weighted mixture of domain properties. The classification of audio segments into domains allows the creation of domain specific acoustic models for automatic speech recognition. Experiments are conducted on a dataset of diverse speech data covering speech from radio and TV broadcasts, telephone conversations, meetings, lectures and read speech, with a joint training set of 60 hours and a test set of 6 hours. Maximum A Posteriori (MAP) adaptation to LDA based domains was shown to yield relative Word Error Rate (WER) improvements of up to 16% relative, compared to pooled training, and up to 10%, compared with models adapted with human-labelled prior domain knowledge.

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。

↑