Unlimited vocabulary speech recognition for agglutinative languages
Open accessAbstract
It is practically impossible to build a word-based lexicon for speech recognition in agglutinative languages that would cover all the relevant words. The problem is that words are generally built by concatenating several prefixes and suffixes to the word roots. Together with compounding and inflections this leads to millions of different, but still frequent word forms. Due to inflections, ambiguity and other phenomena, it is also not trivial to automatically split the words into meaningful parts. Rule-based morphological analyzers can perform this splitting, but due to the handcrafted rules, they also suffer from an out-of-vocabulary problem. In this paper we apply a recently proposed fully automatic and rather language and vocabulary independent way to build sub-word lexica for three different agglutinative languages. We demonstrate the language portability as well by building a successful large vocabulary speech recognizer for each language and show superior recognition performance compared to the corresponding word-based reference systems.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Morfessor has been successfully applied in speech recognition of many agglutinative languages.
“which has been successfully applied in speech recognition of many agglutinative languages [4], [25].”
From Automatic Speech Recognition With Very Large Conversational Finnish and Estonian Vocabularies (2017) · cites this paper as [4]
The Introduction identifies Morfessor as the algorithm used, and the Discussion reports successful speech-recognition results for all three agglutinative languages tested. The paper supports the application to Finnish, Estonian and Turkish, but not the broader claim of application to “many” languages. (LLM, conf=0.82)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
No references could be resolved yet.
Research works printed but not linked: 20
- Cyril Allauzen, Mehryar Mohri, Michael Riley, and Brian Roark. 2004. A generalized construction of integrated speech recognition transducers. In Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Montreal, Canada.
- Tanel Alumäe. 2005. Phonological and morphologi- cal modeling in large vocabulary continuous Estonian speech recognition system. In Proceedings of Second Baltic Conference on Human Language Technologies, pages 89-94.
- Ebru Arisoy and Levent Arslan. 2005. Turkish dictation system for broadcast news applications. In 13th Euro- pean Signal Processing Conference -EUSIPCO 2005, Antalya, Turkey, September.
- Mathias Creutz and Krista Lagus. 2002. Unsupervised discovery of morphemes. In Proceedings of the Work- shop on Morphological and Phonological Learning of ACL-02, pages 21-30.
- J. Garofolo, G. Auzanne, and E. Voorhees. 2000. The TREC spoken document retrieval track: A success story. In Proceedings of Content Based Multimedia Information Access Conference, April 12-14.
- P. Geutner, M. Finke, and P. Scheytt. 1998. Adap- tive vocabularies for transcribing multilingual broad- cast news. In Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Process- ing (ICASSP), Seattle, WA, USA, May.
- H. Erdogan, O. Buyuk, K. Oflazer. 2005. Incorporating language constraints in sub-word based speech recog- nition. IEEE Automatic Speech Recognition and Un- derstanding Workshop, Cancun, Mexico.
- Kadri Hacioglu, Brian Pellom, Tolga Ciloglu, Ozlem Oz- turk, Mikko Kurimo, and Mathias Creutz. 2003. On lexicon creation for Turkish LVCSR. In Proceedings of 8th European Conference on Speech Communica- tion and Technology, pages 1165-1168.
- Jan Kneissler and Dietrich Klakow. 2001. Speech recog- nition for huge vocabularies by using optimized sub- word units. In Proceedings of the 7th European Con- ference on Speech Communication and Technology (Eurospeech), pages 69-72, Aalborg, Denmark.
- Kevin McTait and Martine Adda-Decker. 2003. The 300k LIMSI German Broadcast News Transcription System. In Proceedings of 8th European Conference on Speech Communication and Technology.
- Einar Meister, Jürgen Lasn, and Lya Meister. 2002. Esto- nian SpeechDat: a project in progress. In Proceedings of the Fonetiikan Päivät -Phonetics Symposium 2002 in Finland, pages 21-26.
- NIST. 2000. Proceedings of DARPA workshop on Auto- matic Transcription of Broadcast News. NIST, Wash- ington DC, May.
- Janne Pylkkönen. 2005. New pruning criteria for effi- cient decoding. In Proceedings of 9th European Con- ference on Speech Communication and Technology.
- Janne Pylkkönen and Mikko Kurimo. 2004. Duration modeling techniques for continuous speech recogni- tion. In Proceedings of the International Conference on Spoken Language Processing.
- Murat Saraclar, Michael Riley, Enrico Bocchieri, and Vincent Goffin. 2002. Towards automatic closed cap- tioning: Low latency real time broadcast news tran- scription. In Proceedings of the International Confer- ence on Spoken Language Processing (ICSLP), Den- ver, CO, USA.
- Vesa Siivola and Bryan Pellom. 2005. Growing an n- gram language model. In Proceedings of 9th European Conference on Speech Communication and Technol- ogy.
- Vesa Siivola, Mikko Kurimo, and Krista Lagus. 2001. Large vocabulary statistical language modeling for continuous speech recognition. In Proceedings of 7th European Conference on Speech Communication and Technology, pages 737-747, Aalborg, Copenhagen.
- Andreas Stolcke. 1998. Entropy-based pruning of back- off language models. In Proc. DARPA Broadcast News Transcription and Understanding Workshop, pages 270-274.
- Andreas Stolcke. 2002. SRILM -an extensible lan- guage modeling toolkit. In Proceedings of the Inter- national Conference on Spoken Language Processing, pages 901-904.
- Mate Szarvas and Sadaoki Furui. 2003. Evaluation of the stochastic morphosyntactic language model on a one million word Hungarian task. In Proceedings of the 8th European Conference on Speech Communication and Technology (Eurospeech), pages 2297-2300.
Cited by
Sorted by citations
Automatic Speech Recognition With Very Large Conversational Finnish and Estonian Vocabularies
Enarvi, S., Smit, P., Virpioja, S., & Kurimo, M. (2017) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2017.2743344 · Cited by 33 · see citing works
“We use the Morfessor method [23], [24], which has been successfully applied in speech recognition of many agglutinative languages [4], [25].”
BackgroundMethodSelf-citation
- Authors
- Kurimo, M., Puurula, A., Arisoy, E., Siivola, V., Hirsimäki, T., Pylkkönen, J., Alumäe, T., & Saraclar, M.
- Affiliations
- Tallinn University of Technology (EE)
- Venue
- Proceedings of the main conference on Human Language Technology Conference of the North American Chapter of the Association of Computational Linguistics -
- Published
- 2006
- Pages
- 487-494
- OpenAlex
- W2042783153
- Cited by
- 82
- References
- 0
- Citations by year
- 2024: 1 · 2022: 3 · 2021: 1 · 2020: 6 · 2019: 8 · 2018: 5 · 2017: 4 · 2016: 2 · 2015: 1 · 2014: 6 · 2013: 4 · 2012: 2
- FWCI
- 7.43
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) (publishedVersion)
- Funders
- Academy of Finland
- Türkiye Bilimsel ve Teknolojik Araştırma Kurumu
- China Scholarship Council
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
- Keywords
- agglutinative languages, large vocabulary speech recognition, out-of-vocabulary words, morphological analysis, word segmentation, compounding, inflection, language portability
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
