{ "paper_id": "O12-1030", "header": { "generated_with": "S2ORC 1.0.0", "date_generated": "2023-01-19T08:03:02.365925Z" }, "title": "Implementation and Comparison of Keyword Spotting for Taiwanese", "authors": [ { "first": "Chung-Che", "middle": [], "last": "Wang", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Che-Hsuan", "middle": [], "last": "Chou", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Liang-Yu", "middle": [], "last": "Chen", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Yu-Jhe", "middle": [], "last": "Li", "suffix": "", "affiliation": {}, "email": "liyujhe@mirlab.org" }, { "first": "Jyh-Shing", "middle": [ "Roger" ], "last": "Jang", "suffix": "", "affiliation": {}, "email": "jang@mirlab.org" }, { "first": "Hsun-Cheng", "middle": [], "last": "Hu", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Shih-Peng", "middle": [], "last": "Lin", "suffix": "", "affiliation": {}, "email": "shihpeng@iii.org.tw" }, { "first": "You-Lian", "middle": [], "last": "Huang", "suffix": "", "affiliation": {}, "email": "" } ], "year": "", "venue": null, "identifiers": {}, "abstract": "", "pdf_parse": { "paper_id": "O12-1030", "_pdf_hash": "", "abstract": [], "body_text": [ { "text": "This paper focuses on improving in the performance of a Taiwanese keyword spotting system by integrating speech assessment and pitch contour classification. In the first part of this research, we use different methods to implement a Taiwanese keyword spotting system. In second part, we improve the system by validation using speech assessment and pitch contour classification. Two methods are adopted in the first part to implement the keyword spotting system: hidden Markov model and phone mismatching method. We then perform speech assessment and pitch contour classification to validate the candidate keywords selected by these two methods to refine the results. A threshold is used for a decision tree to make the final decision. Experimental results shows that the equal error rates (ERRs) reduce about 20% and 5% after being incorporated speech assessment validation. After being incorporated with pitch contour classification, ERRs further reduce about 1%. This concludes that the validation technique using speech assessment and pitch contour classification can improve the performance of Taiwanese keyword spotting. 3. ", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Abstract", "sec_num": null }, { "text": "EQUATION", "cite_spans": [], "ref_spans": [], "eq_spans": [ { "start": 0, "end": 8, "text": "EQUATION", "ref_id": "EQREF", "raw_str": "P ( ,", "eq_num": ", , ) DP 1" } ], "section": "4.", "sec_num": null }, { "text": "Proceedings of the Twenty-Fourth Conference on Computational Linguistics and Speech Processing(ROCLING 2012)", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null } ], "back_matter": [], "bib_entries": { "BIBREF0": { "ref_id": "b0", "title": "Estimation of phone mismatch penalty matrices for two-stage keyword spotting", "authors": [ { "first": "C", "middle": [ "W" ], "last": "Han", "suffix": "" }, { "first": "S", "middle": [ "J" ], "last": "Kang", "suffix": "" }, { "first": "N", "middle": [ "S" ], "last": "Kim", "suffix": "" } ], "year": null, "venue": "IEICE TRANSACTIONS on Information and Systems", "volume": "", "issue": "8", "pages": "2331--2335", "other_ids": {}, "num": null, "urls": [], "raw_text": "C. W. Han, S. J. Kang, and N. S. Kim, \"Estimation of phone mismatch penalty matrices for two-stage keyword spotting,\" IEICE TRANSACTIONS on Information and Systems Vol.E93-D No.8 pp.2331-2335", "links": null }, "BIBREF1": { "ref_id": "b1", "title": "Keyword search using modified minimum edit distance measure", "authors": [ { "first": "K", "middle": [], "last": "Audhkhasi", "suffix": "" }, { "first": "A", "middle": [], "last": "Verma", "suffix": "" } ], "year": 2007, "venue": "Proc. ICASSP", "volume": "", "issue": "", "pages": "929--932", "other_ids": {}, "num": null, "urls": [], "raw_text": "K. Audhkhasi and A. Verma, \"Keyword search using modified minimum edit distance measure,\" Proc. ICASSP, pp. 929-932,Apr. 2007.", "links": null }, "BIBREF2": { "ref_id": "b2", "title": "Keyword Spotting based on the Analysis of Template Matching Distances", "authors": [ { "first": "M", "middle": [ "S" ], "last": "Barakat", "suffix": "" }, { "first": "C", "middle": [ "H" ], "last": "Ritz", "suffix": "" }, { "first": "D", "middle": [ "A" ], "last": "Stirling", "suffix": "" } ], "year": 2011, "venue": "5th International Conference on Signal Processing and Communication Systems ICSPCS", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "M. S. Barakat, C. H. Ritz, D. A. Stirling, \"Keyword Spotting based on the Analysis of Template Matching Distances,\" 5th International Conference on Signal Processing and Communication Systems ICSPCS,2011", "links": null }, "BIBREF3": { "ref_id": "b3", "title": "Improved mandarin keyword spotting using confusion garbage model", "authors": [ { "first": "S", "middle": [ "L" ], "last": "Zhang", "suffix": "" }, { "first": "Z", "middle": [ "W" ], "last": "Shuang", "suffix": "" }, { "first": "Q", "middle": [], "last": "Shi", "suffix": "" }, { "first": "Y", "middle": [], "last": "Qin", "suffix": "" } ], "year": 2010, "venue": "Proc. ICPR", "volume": "", "issue": "", "pages": "3700--3703", "other_ids": {}, "num": null, "urls": [], "raw_text": "S.L. Zhang, Z.W. Shuang, Q. Shi, and Y. Qin, \"Improved mandarin keyword spotting using confusion garbage model\", in Proc. ICPR, pp. 3700-3703, Istanbul, Turkey, 2010", "links": null }, "BIBREF4": { "ref_id": "b4", "title": "A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition", "authors": [ { "first": "L", "middle": [ "R" ], "last": "Rabiner", "suffix": "" } ], "year": 1989, "venue": "Proc. Of the IEEE", "volume": "77", "issue": "", "pages": "257--286", "other_ids": {}, "num": null, "urls": [], "raw_text": "L.R. Rabiner, \"A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition\", Proc. Of the IEEE, Vol.77, No.2, pp. 257-286, Feb, 1989.", "links": null }, "BIBREF5": { "ref_id": "b5", "title": "Spoken Language Processing", "authors": [ { "first": "X", "middle": [], "last": "Huang", "suffix": "" }, { "first": "A", "middle": [], "last": "Acero", "suffix": "" }, { "first": "H", "middle": [ "W" ], "last": "Hon", "suffix": "" } ], "year": 2001, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Huang, X., Acero, A., and Hon, H.W., \"Spoken Language Processing\", New Jersey, Prentice Hall, 2001", "links": null }, "BIBREF6": { "ref_id": "b6", "title": "Toward Constructing A Multilingual Speech Corpus for Taiwanese (Min-nan), Hakka, and Mandarin Chinese", "authors": [ { "first": "R.-Y", "middle": [], "last": "Lyu", "suffix": "" }, { "first": "M.-S", "middle": [], "last": "Liang", "suffix": "" }, { "first": "Y.-C", "middle": [], "last": "Chiang", "suffix": "" } ], "year": 2004, "venue": "International Journal of Computational Linguistics & Chinese Language Processing", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "R.-Y. Lyu, M.-S. Liang, Y.-C. Chiang, Toward Constructing A Multilingual Speech Corpus for Taiwanese (Min-nan), Hakka, and Mandarin Chinese, International Journal of Computational Linguistics & Chinese Language Processing, 2004", "links": null }, "BIBREF7": { "ref_id": "b7", "title": "Decision tree based tone modeling with corrective feedbacks for automatic Mandarin tone assessment", "authors": [ { "first": "H.-C", "middle": [], "last": "Liao", "suffix": "" }, { "first": "J.-C", "middle": [], "last": "Chen", "suffix": "" }, { "first": "S.-C", "middle": [], "last": "Chang", "suffix": "" }, { "first": "Y.-H", "middle": [], "last": "Guan", "suffix": "" }, { "first": "C.-H", "middle": [], "last": "Lee", "suffix": "" } ], "year": null, "venue": "INTERSPEECH 2010", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Liao, H.-C., Chen, J.-C. , Chang, S.-C., Guan, Y.-H., Lee, C.-H., \"Decision tree based tone modeling with corrective feedbacks for automatic Mandarin tone assessment\", In INTERSPEECH 2010.", "links": null }, "BIBREF8": { "ref_id": "b8", "title": "TRUES: Tone Recognition Using Extended Segment", "authors": [ { "first": "J.-C", "middle": [], "last": "Chen", "suffix": "" }, { "first": "J", "middle": [ "S R" ], "last": "Jang", "suffix": "" } ], "year": 2008, "venue": "ACM Trans. Asian Lang. Inform. Process", "volume": "7", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Chen, J.-C., Jang, J.S. R., \"TRUES: Tone Recognition Using Extended Segment\", ACM Trans. Asian Lang. Inform. Process. 7, 3, Article 10, 2008.", "links": null }, "BIBREF9": { "ref_id": "b9", "title": "Average Magnitude Difference Function Pitch Extractor", "authors": [ { "first": "M", "middle": [], "last": "Ross", "suffix": "" }, { "first": "H", "middle": [], "last": "Shaffer", "suffix": "" }, { "first": "A", "middle": [], "last": "Cohen", "suffix": "" }, { "first": "R", "middle": [], "last": "Freudberg", "suffix": "" }, { "first": "H", "middle": [], "last": "Manley", "suffix": "" } ], "year": 1974, "venue": "IEEE Transactions on Acoustics, Speech, and Signal Processing", "volume": "22", "issue": "5", "pages": "353--362", "other_ids": {}, "num": null, "urls": [], "raw_text": "Ross, M.Shaffer, H. Cohen, A. Freudberg, R., and Manley, H., \"Average Magnitude Difference Function Pitch Extractor,\" IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol. 22,No. 5,353-362, 1974", "links": null } }, "ref_entries": { "TABREF1": { "text": "Proceedings of the Twenty-Fourth Conference on Computational Linguistics and Speech Processing(ROCLING 2012)", "html": null, "num": null, "type_str": "table", "content": "
indicatorlog
[17]
()
Pr
filler-model
2.
[2]
1861
fi_initfi_final
iI D[2]
Keywords: Keywords spotting, hidden Markov model, penalty matrix
3C
3C
[1]
Phone Mismatch Penalty Matrixconfidence measure, CM development
substitutioninsertion
deletion
[1]
1.
" }, "TABREF2": { "text": "Proceedings of the Twenty-Fourth Conference on Computational Linguistics and Speech Processing(ROCLING 2012)", "html": null, "num": null, "type_str": "table", "content": "
Proceedings of the Twenty-Fourth Conference on Computational Linguistics and Speech Processing (ROCLING 2012) Proceedings of the Twenty-Fourth Conference on Computational Linguistics and Speech Processing (ROCLING 2012) Proceedings of the Twenty-Fourth Conference on Computational Linguistics and Speech Processing (ROCLING 2012) Proceedings of the Twenty-Fourth Conference on Computational Linguistics and Speech Processing (ROCLING 2012) Proceedings of the Twenty-Fourth Conference on Computational Linguistics and Speech Processing (ROCLING 2012) Proceedings of the Twenty-Fourth Conference on Computational Linguistics and Speech Processing (ROCLING 2012)
1.(70log-likelihood 97 539 692 533 2 ) \u03d2 70 HMM phone sequence14 GMM (3)
2. FAR\u03d2 rank1.8%46.5%a b\u03d226.5% a bp
2.
34% 39.4%28.4% 34.6%
HMM27.3% 33.7%decision
tree
3.[11]
3.7[14]-p -t -k -h1715896
[4]
3.5 4 4.554.5
HMMDET
4. 5.6% false acceptance rate , FAR [16] HMM likelihood ratio testing 4.8% baseline durationfalse rejection rate ,FRR [16] equal error rate, EER [16] ( ) HMM 20% time duration baselineLRT pause
1
420%
[7]7050development
4.log-likelihoodrank ratio score -8
1. ForPArank ratio score [12][13] penalty -20 NSC 99-2221-E-007 -049 -MY3 50.32%confidence measure
D (P , programming, DP (development) 1.8%) UPDUDP[9]UPDUDP[9] FRRFRRFAR Dynamic
AMDF[10] (average magnitude difference function ) 34%FARFARDP
P1.1%log-likelihood HTKQ 0.9%
(n) C 0 1niirow[8]Gaussian mixture model, HMM
GMMcolumn[15]
(2) (3) HMM
2.backtracking
(2)
1.ForPAlau2_heh7 ForPA14
" } } } }