고립 단어 인식 결과의 비유사 후보 단어 제외 성능을 개선하기 위한 다양한 접근 방법 연구
Various Approaches to Improve Exclusion Performance of Non-similar Candidates from N-best Recognition Results on Isolated Word Recognition
3
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Many isolated word recognition systems may generate non-similar words for recognition candidates because they use only acoustic information. The previous study [1,2] investigated several techniques which can exclude non-similar words from N-best candidate words by applying Levenstein distance measure. This paper discusses the various improving techniques of removing non-similar recognition results. The mentioned methods include comparison penalties or weights, phone accuracy based on confusion information, weights candidates by ranking order and partial comparisons. Through experimental results, it is found that some proposed method keeps more accurate recognition results than the previous method's results.
Abstract
Many isolated word recognition systems may generate non-similar words for recognition candidates because they use only acoustic information. The previous study [1,2] investigated several techniques which can exclude non-similar words from N-best candidate words by applying Levenstein distance measure. This paper discusses the various improving techniques of removing non-similar recognition results. The mentioned methods include comparison penalties or weights, phone accuracy based on confusion information, weights candidates by ranking order and partial comparisons. Through experimental results, it is found that some proposed method keeps more accurate recognition results than the previous method's results.
참고문헌 (5)
- 윤영선, "고립단어 인식 시스템의 후보 단어열에서 비유사 후보단어의 제외 연구", 2008 대한음성학회·한국음성학회 공동학술대회 발표논문집, pp.183~186 (2008)
- 윤영선, "레벤스타인 거리에 기초한 위치 정확도를 이용한 고립 단어 인식 결과의 비유사 후보 단어 제외", 말소리와 음성과학, 1(3), pp.109~115 (2009), 한국음성학회
- 박전규, "내장형 대어휘 음성인식 기술에 기반하는 행선지 입력 시스템 개발", 2007 대한음성학회 가을학술대회 발표논문집, pp.108~111 (2007)
- Ueffing, N., "Confidence measures for statistical machine translation", Proceedings of MT Summit IX (2003)
- "Sequence Alignment"[URL]