摘要
解码器引擎是语音识别系统的核心模块,而基于加权有限状态机(WFST)的解码器则是解码器的一种典型形式。分析静态WFST解码器在实际应用中的资源占用问题,提出一种在解码和词图生成过程中通过检测非活跃节点动态回收系统资源的策略。最后,在OpenKWS 15数据集上进行实验,证明该策略使解码器的内存消耗比不回收系统资源的解码器降低75%左右。
Decoder is the core module of speech recognition system, and the decoder based on the weighted finite-state transducers(WFST) is a typical form of decoder. We analyze the resource occupation of WFST-based static decoder in practice, and propose a strategy for dynamical recovery of system resources by detecting inactive nodes during decoding and lattice generation. Finally, we carry out experiments on the OpenKWS 15 dataset to show that the decoder with this strategy consumes about 75% less memory than decoders that do not reclaim system resources.
作者
丁佳伟
刘加
张卫强
冯运波
刘利军
于乐
DING Jiawei;LIU Jia;ZHANG Weiqiang;FENG Yunbo;LIU Lijun;YU Le(Department of Electronic Engineering,Tsinghua University,Beijing 100084,China;China Mobile Information Security Center,Beijing 100053,China)
出处
《中国科学院大学学报(中英文)》
CSCD
北大核心
2019年第1期109-114,共6页
Journal of University of Chinese Academy of Sciences
基金
国家自然科学基金(U1836219)资助