首页|Language Model Score Regularization for Speech Recognition

Language Model Score Regularization for Speech Recognition

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
  • 维普
Inspired by the fact that back-off and interpolated smoothing algorithms have significant effect on statistical language modeling, this paper proposes a sentence-level Language model (LM) score regularization algorithm to improve the fault-tolerance of LMs for recognition errors. The proposed algorithm is applicable to both count-based LMs and neural network LMs. Instead of predicting the occurrence of a sequence of words under a fixed order Markov assumption, we use a composite model consisting of different order models with either n-gram or skip-gram features to estimate the probability of the sequence of words. In order to simplify implementations, we derive a connection between bidirectional neural networks and the proposed algorithm. Experiments were carried out on the Switchboard corpus. Results on N-best lists re-scoring show that the proposed algorithm achieves consistent word error rate reduction when it is applied to count-based LMs, Feedforward neural network (FNN) LMs, and Recurrent neural network (RNN) LMs.

Speech recognitionLanguage model score regularizationInterpolationBidirectional neural network

ZHANG Yike、ZHANG Pengyuan、YAN Yonghong

展开 >

Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190, China

University of Chinese Academy of Sciences, Beijing 100049, China

Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumchi 830011, China

National Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaNational Key Research and Development PlanNational Key Research and Development PlanKey Science and Technology Project of the Xinjiang Uygur Autonomous RegionPre-research Project for Equipment of General Information System

U153611711590770-42016YFB08012032016YFB08012002016A03007-1JZX2017-0994/Y306

2019

10.1049/cje.2019.03.015

中国电子杂志(英文版)

中国电子杂志(英文版)

CSTPCDCSCDSCIEI
ISSN：1022-4653
年,卷(期)：2019.28(3)
  • 26