Importance of high-order N-gram models in morph-based speech recognition

IEEE Transactions on Audio, Speech and Language Processing

Volumn 17, Issue 4, 2009, Pages 724-732

  Hirsimaki, Teemu ^a   Pylkkonen, Janne ^a   Kurimo, Mikko ^a  

^a AALTO UNIVERSITY   (Finland)

Author keywords

Language modeling (LM); Morphology; Speech recognition; Variable length n grams

Indexed keywords

AGGLUTINATIVE LANGUAGES; FINNISH; HIGH ORDERS; LANGUAGE MODELING (LM); LANGUAGE MODELS; LARGE VOCABULARIES; N-GRAM MODELS; RECOGNITION ERRORS; SPEECH RECOGNITION SYSTEMS; VARIABLE-LENGTH N-GRAMS; WORD-BASED MODELS;

COMPUTATIONAL LINGUISTICS; ERROR DETECTION; FACE RECOGNITION; LAGRANGE MULTIPLIERS; NATURAL LANGUAGE PROCESSING SYSTEMS; QUERY LANGUAGES; SPEECH ANALYSIS; VOCABULARY CONTROL;

SPEECH RECOGNITION;

EID: 65349113250     PISSN: 15587916     EISSN: None     Source Type: Journal    
DOI: 10.1109/TASL.2008.2012323     Document Type: Article

References (37)

1. G. Choueiter, D. Povey, S. F. Chen, and G. Zweig, "Morpheme-based language modeling for Arabic LVCSR," in Proc. Int. Conf. Acoust., Speech Signal Process., 2006, pp. I-1053-I-1056.
2. B. Xiang, K. Nguyen, L. Nguyen, R. Schwartz, and J. Makhoul, "Morphological decomposition for Arabic broadcast news transcription," in Proc. Int. Conf. Acoust., Speech Signal Process., 2006, pp. I-1089-I-1092.
3. J. Nouza, J. Ždánský, P. David, P. Červa, J. Kolorenč, and D. Nejedlová, "Fully automated system for Czech spoken broadcast transcription with very large (300 k+) lexicon," in Proc. Interspeech, 2005, pp. 1681-1684.
4. T. Hirsimäki, M. Creutz, V. Siivola, M. Kurimo, S. Virpioja, and J. Pylkkönen, "Unlimited vocabulary speech recognition with morph language models applied to Finnish," Comput. Speech Lang., vol. 20, no. 4, pp. 515-541, Oct. 2006.
5. K. McTait and M. Adda-Decker, "The 300 k LIMSI German broadcast news transcription system," in Proc. Eurospeech, 2003, pp. 213-216.
6. W. J. Byrne, J. Hajič, P. Krbec, P. Ircing, and J. Psutka, "Morpheme based language models for speech recognition of Czech," in Proc. 3rd Int. Workshop Text, Speech, Dialogue (TSD), 2000, pp. 211-216.
7. T. Rotovnik, M. S. Maučec, and Z. Kačič, "Large vocabulary continuous speech recognition of an inflected language using stems and endings," Speech Commun., vol. 49, no. 6, pp. 437-452, Jun. 2007.
8. H. Erdogan, O. Büyük, and K. Oflazer, "Incorporating language constraints in sub-word based speech recognition," in Proc. IEEE Workshop Automat. Speech Recognition Understanding (ASRU), 2005, pp. 98-103.
9. M. Creutz, T. Hirsimäki, M. Kurimo, A. Puurula, J. Pylkkönen, V. Siivola, M. Varjokallio, E. Arisoy, M. Saraçlar, and A. Stolcke, "Morphbased speech recognition and modeling of out-of-vocabulary words across languages," ACM Trans. Speech Lang. Process., vol. 5, no. 1, Dec. 2007.
10. K. Kirchhoff, D. Vergyri, J. Bilmes, K. Duh, and A. Stolcke, "Morphology-based language modeling for conversational Arabic speech recognition," Comput. Speech Lang., vol. 20, no. 4, pp. 589-608, Oct. 2006.
11. R. Sarikaya, M. Afify, and Y. Gao, "Joint morphological-lexical language modeling (JMLLM) for Arabic," in Proc. Int. Conf. Acoust., Speech Signal Process., 2007, pp. IV-181-IV-184.
12. W. Byrne, J. Hajič, P. Ircing, F. Jelinek, S. Khudanpur, P. Krbec, and J. Psutka, "On large vocabulary continuous speech recognition of highly inflectional language-Czech," in Proc. Eurospeech, 2001, pp. 487-489.
13. T. Laureys, V. Vandeghinste, and J. Duchateau, "A hybrid approach to compounds in LVCSR," in Proc. Int. Conf. Spoken Lang. Process., 2002, pp. 697-700.
14. T. Alumäe, "Methods for Estonian large vocabulary speech recognition," Ph.D. dissertation, Tallinn Univ. of Technology, Tallinn, Estonia, 2006.
15. M. Kurimo, A. Puurula, E. Arisoy, V. Siivola, T. Hirsimäki, J. Pylkkönen, T. Alumäe, and M. Saraçlar, "Unlimited vocabulary speech recognition for agglutinative languages," in Proc. Human Lang. Technol. Conf. North Amer. Chapter Assoc. Comput. Linguist., 2006, pp. 487-494.
16. A. Puurula and M. Kurimo, "Vocabulary decomposition for Estonian open vocabulary speech recognition," in Proc. 45th Annu. Meeting Assoc. Comput. Linguist., 2007, pp. 89-95.
17. V. Siivola, T. Hirsimäki, M. Creutz, and M. Kurimo, "Unlimited vocabulary speech recognition based on morphs discovered in an unsupervised manner," in Proc. Eurospeech, 2003, pp. 2293-2296.
18. P. Geutner, "Using morphology towards better large-vocabulary speech recognition systems," in Proc. Int. Conf. Acoust., Speech Signal Process., 1995, pp. 445-448.
19. P. Mihajlik, T. Fegyó, Z. Tüske, and P. Ircing, "A morpho-graphemic approach for the recognition of spontaneous speech in agglutinative languages-Like Hungarian," in Proc. Interspeech, 2007, pp. 1497-1500.
20. M. Szarvas and S. Furui, "Evaluation of the stochastic morphosyntactic language model on a one million word Hungarian task," in Proc. Eurospeech, 2003, pp. 2297-2300.
21. M. S. Maučec, T. Rotovnik, and M. Zemljak, "Modelling highly inflected Slovenian language," Int. J. Speech Technol., vol. 6, no. 3, pp. 245-257, Jul. 2003.
22. K. Hacioglu, B. Pellom, T. Ciloglu, O. Ozturk, M. Kurimo, and M. Creutz, "On lexicon creation for Turkish LVCSR," in Proc. Eurospeech, 2003, pp. 1165-1168.
23. E. Arisoy, H. Dutaǧaci, and L. M. Arslan, "A unified language model for large vocabulary continuous speech recognition of Turkish," Signal Process., vol. 86, no. 10, pp. 2844-2862, Oct. 2006.
24. E. Arisoy, H. Sak, and M. Saraçlar, "Language modeling for automatic Turkish broadcast news transcription," in Proc. Interspeech, 2007, pp. 2381-2384.
25. M. Kurimo, M. Creutz, M. Varjokallio, E. Arisoy, and M. Saraçlar, "Unsupervised segmentation ofwords into morphemes-Morpho challenge 2005: Application to automatic speech recognition," in Proc. Int. Conf. Spoken Lang. Process., 2006, pp. 1021-1024.
26. M. Kurimo, M. Creutz, and M. Varjokallio, "Morpho challenge evaluation using a linguistic gold standard," in Proc. CLEF 2007 workshop, 2008, vol. 5152. Springer, pp. 864-872.
27. S. F. Chen and J. Goodman, "An empirical study of smoothing techniques for language modeling," Comput. Speech Lang., vol. 13, no. 4, pp. 359-393, Oct. 1999.
28. A. Stolcke, "Entropy-based pruning of backoff language models," in Proc.DARPA Broadcast News Transcription and Understanding Workshop, 1998, pp. 270-274.
29. V. Siivola, T. Hirsimäki, and S. Virpioja, "On growing and pruning Kneser-Ney smoothed n-gram models," IEEE Trans. Audio, Speech Lang. Process., vol. 15, no. 5, pp. 1617-1624, Jul. 2007.
30. M. Creutz and K. Lagus, Unsupervised Morpheme Segmentation and Morphology Induction From Text Corpora Using Morfessor 1.0. Espoo, Finland: Helsinki Univ. of Technology, Publications in Computer and Information Science A81, 2005.
31. M. Mohri, F. Pereira, and M. Riley, "Weighted finite-state transducers in speech recognition," Comput. Speech Lang., vol. 16, no. 1, pp. 69-88, Jan. 2002.
32. J. Pylkkönen, "An efficient one-pass decoder for Finnish large vocabulary continuous speech recognition," in Proc. 2nd Baltic Conf. Human Lang. Technol., 2005, pp. 167-172.
33. S. J. Young, N. H. Russell, and J. H. S. Thornton, "Token passing: A simple conceptual model for connected speech recognition systems," Cambridge Univ. Eng. Dept., 1989, Technical report.
34. S. Ortmanns, H. Ney, and A. Eiden, "Language-model look-ahead for large vocabulary speech recognition," in Proc. Int. Conf. Spoken Lang. Process., 1996, pp. 2095-2098.
35. X. Aubert and H. Ney, "Large vocabulary continuous speech recognition using word graphs," in Proc. Int. Conf. Acoust., Speech Signal Process., 1995, pp. 49-52.
36. Finnish text collection 2004 [Online]. Available: http://www.csc.fi/ kielipankki/
37. E. Meister, J. Lasn, and L. Meister, "Development of the Estonian SpeechDat-like database," in Proc. Eurospeech, 2003, pp. 1601-1604.