한국음향학회지 (The Journal of the Acoustical Society of Korea)

  • 제23권5호
  • /
  • Pages.410-419
  • /
  • 2004
  • /
  • 1225-4428(pISSN)
  • /
  • 2287-3775(eISSN)
한국음향학회 (The Acoustical Society of Korea)
권호 표지

베이시안 신뢰도 융합을 이용한 신뢰도 측정

Bayesian Fusion of Confidence Measures for Confidence Scoring

  • 김태윤 (고려대학교 전자컴퓨터공학과) ;
  • 고한석 (고려대학교 전자컴퓨터공학과)
  • 발행 : 2004.07.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 베이시안에 기반한 신뢰도 융합 기법을 제안한다. 음성인식에서 신뢰도는 인식 결과에 대한 신뢰의 정도를 말하며, 인식 결과가 맞는 지의 여부를 판단할 수 있다. 개별 신뢰도 기법의 신뢰도 값을 융합하여 최종 판단을 내리는 집중형 융합 방식과 개별 신뢰도 기법의 판단 결과들을 융합하는 분산형 융합의 두 가지 방식에 대해 최적의 베이시안 융합규칙이 제시되었다. 고립단어 인식에서의 미등록어 거절 실험 결과 집중형 베이시안 신뢰도 융합 기법은 개별 신뢰도 기법에 비해 13% 이상의 상대적인 에러 감소 효과를 보였으나, 분산형 베이시안 융합은 성능의 향상을 보이지 못했다.

In this paper. we propose a method of confidence measure fusion under Bayesian framework for speech recognition. Centralized and distributed schemes are considered for confidence measure fusion. Centralized fusion is feature level fusion which combines the values of individual confidence scores and makes a final decision. In contrast. distributed fusion is decision level fusion which combines the individual decision makings made by each individual confidence measuring method. Optimal Bayesian fusion rules for centralized and distributed cases are presented. In isolated word Out-of-Vocabulary (OOV) rejection experiments. centralized Bayesian fusion shows over 13% relative equal error rate (EER) reduction compared with the individual confidence measure methods. In contrast. the distributed Bayesian fusion shows no significant performance increase.

키워드

참고문헌

  1. Proc. EUROSPEECH Is This Conversation on Track? Paul Carpenter;Chun Jin(et al.)
  2. Proc. ICSLP Improved MLLR Speaker Adaptation Using Confidence Measures for Conversation Speech Recognition Micheal Pitz;Frank Wessel(et al.)
  3. Proc. EUROSPEECH Word Lavel Confidence Annotation using Combinations of Features Rong Zhang;Alexander Rudnicky
  4. IEEE Trans. on Speech and Audio Processing v.9 no.3 Confidence Measures for Large Vocabulary Continuous Speech Recognition Frank Wessel;Schluter(et al.) https://doi.org/10.1109/89.906002
  5. IEEE Trans. on Speech and Audio Processing v.8 no.2 Utterance Verification in Continuous Speech Recognition: Decoding and Training Procedures Eduardo Lleida;Richard C. Rose https://doi.org/10.1109/89.824697
  6. IEEE Trans. on Speech and Audio Processing v.4 no.6 Vocabulary Independent Discriminative Utterance Verification for Nonkeyword Rejection in Subword Based Speech Recognition Rafid A. Sukkar;Chin-Hui Lee https://doi.org/10.1109/89.544527
  7. proc. ICASSP v.3 Word and Phone Level Acoustic Confidence Scoring Simo Kamppari;Timothy Hazen
  8. proc. ICASSP Neural-Network Based Measures of Confidence for Word Recognition Mitch Weintraub;Francoise Beaufays(et al.)
  9. proc. EUROSPEECH A Boosting Approach for Confidence Scoring Pedro J. Moreno;Beth Logan(et al.)
  10. Proceedings of the IEEE v.85 no.1 An Introduction to Multisensor Data Fusion David L. Hall;James Llians https://doi.org/10.1109/5.554205
  11. Proceedings of the IEEE v.85 no.1 Distributed Detection with Multiple Sensors: Part 1 - Fundamentals Ramanarayanan Viswanathan;Pramod Varshney https://doi.org/10.1109/5.554208
  12. Distributed Detection and Data Fusion Pramod Varshney
  13. IEEE Trans. on Sys., Man. and Cybernetics v.21 no.1 An Algorithm for Datermining the Decision Thresholds in a Distributed Detection Problem Zhunag-Bo Tang;Krishna R. Pattipati(et al.) https://doi.org/10.1109/21.101153
  14. Introduction to Statistical Pattern Recognition(2nd edition) Fukunaga