지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술대회자료
- 저자정보
- 저널정보
- 한국정보통신학회 INTERNATIONAL CONFERENCE ON FUTURE INFORMATION & COMMUNICATION ENGINEERING 2014 INTERNATIONAL CONFERENCE ON FUTURE INFORMATION & COMMUNICATION ENGINEERING Vo.6 No.1
- 발행연도
- 2014.6
- 수록면
- 69 - 72 (4page)
이용수
초록· 키워드
오류제보하기
This research proposed 2-pass sound source indexing method to apply to audio surveillance system, and developed the simulation being applied it. The proposed indexing method is verified the performance by developing simulation tool. The sound source indexing is verified the performance by setting 10 indexing class and making 3 GMM model per class. More detailed 2- pass indexing was done through improving the classification algorithm applied to the existing audio indexing and searching system and adding the sound source recognition process by GMM. The result was more improved than 1-pass classification indexing method by characteristic parameter analysis.
목차
Abstract
I. INTRODUCTION
II. SPEECH CHARACTERISTICS PARAMETERS
III. THE SOUND SOURCE INDEXING PART
IV. DISCUSSION AND CONCLUSIONS
REFERENCES
참고문헌 (0)
참고문헌 신청이 논문의 저자 정보
최근 본 자료
댓글(0)
0
UCI(KEPA) : I410-ECN-0101-2018-004-000962157