한국어 음성인식 플랫폼(ECHOS)의 개선 및 평가

Abstract

이 논문은 2005년도 음성정보기술산업지원센터(SiTEC)의 연구비 지원에 의하여 연구되었음.†이 논문은 2005년도 충북대학교 학술연구지원사업의 연구비 지원에 의하여 연구되었음.攀攀Improvement and Evaluation of the Korean Large Vocabulary Continuous Speech Recognition Platform (ECHOS)Oh-Wook Kwon, Sukbong Kwon, Sungrack Yun, Gyucheol Jang, Yong-Rae Kim, Bong-Wan Kim, Hoirin Kim, Changdong Yoo, Yong-Ju LeeWe report the evaluation results of the Korean speech recognition platform called ECHOS. The platform has an object-oriented and reusable architecture so that researchers can easily evaluate their own algorithms. The platform has all intrinsic modules to build a large vocabulary speech recognizer: Noise reduction, end-point detection, feature extraction, hidden Markov model (HMM)-based acoustic modeling, cross-word modeling, n-gram language modeling, n-best search, word graph generation, and Korean-specific language processing. The platform supports both lexical search trees and finite-state networks. It performs word-dependent n-best search with bigram in the forward search stage, and rescores the lattice with trigram in the backward stage. In an 8000-word continuous speech recognition task, the platform with a lexical tree increases 40% of word errors but decreases 50% of recognition time compared to the HTK platform with flat lexicon. ECHOS reduces 40% of recognition errors through incorporation of cross-word modeling. With the number of Gaussian mixtures increasing to 16, it yields word accuracy comparable to the previous lexical tree-based platform, Julius.