Abstract
Whenever an ASR company promises to deliver error-proof transcripts to the end user, manual verification and correction of the raw ASR transcripts cannot be avoided. This manual post-editing process systematically generates new and correct domain-specific data which can be used to incrementally improve the original ASR system. This paper proposes a statistic, SMT-based ASR error correction method, which takes advantage of the past corrected ASR errors to automatically post-process its future transcripts. We show that the proposed method can bring more than 10% WER improvements using only 2000 user-corrected sentences.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Yu, D., Hwang, M., Mau, P., Acero, A., Deng, L.: Unsupervised Learning from Users’ Error Correction in Speech Dictation. In: 8th International Conference on Spoken Language Processing (Interspeech), Jeju Island, Korea, pp. 1969–1972 (2004)
Cucu, H., Besacier, L., Burileanu, C., Buzo, A.: Investigating the Role of Machine Translated Text in ASR Domain Adaptation: Unsupervised and Semi-supervised Methods. In: The 2011 Automatic Speech Recognition and Understanding Workshop (ASRU 2011), Hawaii, USA, pp. 260–265 (2011)
Cucu, H., Besacier, L., Burileanu, C., Buzo, A.: ASR Domain Adaptation Methods for Low-Resourced Languages: Application to Romanian Language. In: 20 th European Signal Processing Conference (EUSIPCO), Bucharest, Romania (2012)
Ringger, E.K., Allen, J.F.: A Fertility Channel Model for Post-Correction of Continuous Speech Recognition. In: 4th International Conference on Spoken Language Processing, Philadelphia, USA, vol. 2, pp. 897–900 (1996)
Jung, S., Jeong, M., Lee, G.G.: Speech recognition error correction using maximum entropy language model. In: 8th International Conference on Spoken Language Processing (Interspeech), Jeju Island, Korea, pp. 2137–2140 (2004)
Kaki, S., Sumita, E., Iida, H.: A method for correcting errors in speech recognition, using the statistical features of character co-occurrence. In: 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics (COLING-ACL), Montreal, Canada, pp. 653–657 (1998)
Brandow, R.L., Strzalkowski, T.: Improving speech recognition through text-based linguistic post-processing. United States Patent 6064957 (2000)
Mangu, L., Padmanabhan, M.: Error corrective mechanisms for speech recognition. In: 27th International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Salt Lake City, USA, vol. 1, pp. 29–32 (2001)
Sarma, A., Palmer, D.D.: Context-based speech recognition error detection and correction. In: Human Language Technologies: The Annual Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL 2004), Boston, USA, pp. 85–88 (2004)
Moses Statistical Machine Translation Toolkit, http://www.statmt.org/moses
Simard, M., Ueffing, N., Isabelle, P., Kuhn, R.: Rule-Based Translation with Statistical Phrase-Based Post-Editing. In: 2nd Workshop on Statistical Machine Translation, Prague, Czech Republic, pp. 203–206 (2007)
Simard, M., Goutte, C., Isabelle, P.: Statistical Phrase-based Post-editing. In: Human Language Technologies: The Annual Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL 2007), Rochester, NY, USA, pp. 508–515 (2007)
Lagarda, A.L., Alabau, V., Casacuberta, F., Silva, R., Díaz-de-Liaño, E.: Statistical Post-Editing of a Rule-Based Machine Translation System. In: Human Language Technologies: The Annual Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL 2009), Boulder, CO, USA, pp. 217–220 (2009)
Cucu, H., Besacier, L., Burileanu, C., Buzo, A.: Enhancing Automatic Speech Recognition for Romanian by Using Machine Translated and Web-based Text Corpora. In: The 15th International Conference SPECOM, Kazan, Russia (2011)
CMU-Sphinx Speech Recognition Toolkit, http://cmusphinx.sourceforge.net
SRI Language Modeling Toolkit, http://www.speech.sri.com/projects/srilm
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Cucu, H., Buzo, A., Besacier, L., Burileanu, C. (2013). Statistical Error Correction Methods for Domain-Specific ASR Systems. In: Dediu, AH., Martín-Vide, C., Mitkov, R., Truthe, B. (eds) Statistical Language and Speech Processing. SLSP 2013. Lecture Notes in Computer Science(), vol 7978. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-39593-2_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-39593-2_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-39592-5
Online ISBN: 978-3-642-39593-2
eBook Packages: Computer ScienceComputer Science (R0)