Unsupervised Lexicon Discovery from Acoustic Input

Unsupervised Lexicon Discovery from Acoustic Input
复制标题

从声音输入中进行无监督词典发现

DOI:
10.1162/tacl_a_00146
复制
发表时间:
2015
影响因子:
10.9
通讯作者:
James R. Glass
James R. Glass
中科院分区:
人文科学1区
文献类型:
--
作者:
Chia;Timothy J. O'Donnell;James R. Glass

文献摘要

参考文献

被引文献

相似文献

我们提出了一个无监督音位词汇发现模型——从声学输入中同时学习类音位和类词单位的问题。我们的模型建立在早期的声学数据无监督类电话单元发现模型(Lee and Glass, 2012)和使用Adaptor语法框架的无监督符号词汇发现模型(Johnson et al., 2006)的基础上,使用语音变化的概率模型整合了这些早期的方法。我们证明了该模型与最先进的口语术语发现系统具有竞争力,并提出了探索模型行为和它学习的语言结构类型的分析。
We present a model of unsupervised phonological lexicon discovery—the problem of simultaneously learning phoneme-like and word-like units from acoustic input. Our model builds on earlier models of unsupervised phone-like unit discovery from acoustic data (Lee and Glass, 2012), and unsupervised symbolic lexicon discovery using the Adaptor Grammar framework (Johnson et al., 2006), integrating these earlier approaches using a probabilistic model of phonological variation. We show that the model is competitive with state-of-the-art spoken term discovery systems, and present analyses exploring the model’s behavior and the kinds of linguistic structures it learns.
DOI: --
发表时间: --
期刊: --
影响因子: --
作者:
Micha Elsner (Author)
通讯作者: Micha Elsner (Author)