Abstract
Cross-language speech retrieval systems face a cascade of errors due to transcription and translation ambiguity. Using 1-best speech recognition and 1-best translation in such a scenario could adversely affect recall if those 1-best system guesses are not correct. Accurately representing transcription and translation probabilities could therefore improve recall, although possibly at some cost in precision. The difficulty of the task is exacerbated when working with languages for which limited resources are available, since both recognition and translation probabilities may be less accurate in such cases. This paper explores the combination of expected term counts from recognition with expected term counts from translation to perform cross-language speech retrieval in which the queries are in English and the spoken content to be retrieved is in Tagalog or Swahili. Experiments were conducted using two query types, one focused on term presence and the other focused on topical retrieval. Overall, the results show that significant improvements in ranking quality result from modeling transcription and recognition ambiguity, even in lower-resource settings, and that adapting the ranking model to specific query types can yield further improvements.
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
Notes
- 1.
Material is an acronym for Machine Translation for English Retrieval of Information in Any Language [21].
- 2.
In the MATERIAL program these are referred to as conceptual and simple queries, but we prefer to refer to them as topical and lexical in keeping with the way those terms are used in information retrieval and natural language processing, respectively. Some topical and lexical queries also contain additional clues (e.g., synonyms or hypernyms) to guide the interpretation of query terms, but we do not make use of these additional clues in our experiments.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
References
Can, D., Saraclar, M.: Lattice indexing for spoken term detection. IEEE Trans. Audio Speech Lang. Process. 19(8), 2338–2347 (2011)
Chelba, C., et al.: Retrieval and browsing of spoken content. IEEE Signal Process. Mag. 25(3), 39–49 (2008)
Chen, G., et al.: Using proxies for OOV keywords in the keyword search task. In: ASRU, pp. 416–421 (2013)
Darwish, K., Oard, D.: Probabilistic structured query methods. In: SIGIR, pp. 338–344 (2003)
Fiscus, J., Doddington, G.: Topic detection and tracking evaluation overview. In: Allan, J. (ed.) Topic Detection and Tracking. The Information Retrieval Series, vol. 12, pp. 17–31. Springer, Boston (2002)
Hull, D.: Using structured queries for disambiguation in cross-language information retrieval. In: AAAI Symposium on Cross-Language Text and Speech Retrieval (1997)
Karakos, D., et al.: Score normalization and system combination for improved keyword spotting. In: ASRU, pp. 210–215 (2013)
Kim, S., et al.: Combining lexical and statistical translation evidence for cross-language information retrieval. JASIST 66(1), 23–39 (2015)
Lee, L.S., Chen, B.: Spoken document understanding and organization. IEEE Signal Process. Mag. 22(5), 42–60 (2005)
Lee, L.S., Pan, Y.C.: Voice-based information retrieval—how far are we from the text-based information retrieval? In: ASRU, pp. 26–43 (2009)
Makhoul, J., et al.: Speech and language technologies for audio indexing and retrieval. Proc. IEEE 88(8), 1338–1353 (2000)
Mamou, J., et al.: Developing keyword search under the IARPA Babel program. In: Afeka Speech Processing Conference (2013)
McNamee, P., Mayfield, J.: Comparing cross-language query expansion techniques by degrading translation resources. In: SIGIR, pp. 159–166 (2002)
Oard, D.W., et al.: Overview of the CLEF-2006 cross-language speech retrieval track. In: Peters, C., et al. (eds.) CLEF 2006. LNCS, vol. 4730, pp. 744–758. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-74999-8_94
Och, F., Ney, H.: A systematic comparison of various statistical alignment models. Comput. Linguist. 29(1), 19–51 (2003)
Pecina, P., Hoffmannová, P., Jones, G.J.F., Zhang, Y., Oard, D.W.: Overview of the CLEF-2007 cross-language speech retrieval track. In: Peters, C., et al. (eds.) CLEF 2007. LNCS, vol. 5152, pp. 674–686. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-85760-0_86
Pirkola, A.: The effects of query structure and dictionary setups in dictionary-based cross-language information retrieval. In: SIGIR, pp. 55–63 (1998)
Ragni, A., Gales, M.: Automatic speech recognition system development in the ‘wild’. In: ICSA, pp. 2217–2221 (2018)
Riedhammer, K., et al.: A study on LVCSR and keyword search for tagalog. In: INTERSPEECH, pp. 2529–2533 (2013)
Robertson, S.: Okapi at TREC-7: automatic ad hoc, filtering, VLC and interactive track. In: TREC (1998)
Rubino, C.: IARPA MATERIAL program (2016). https://www.iarpa.gov/index.php/research-programs/material/material-baa
Saraclar, M., Sproat, R.: Lattice-based search for spoken utterance retrieval. In: NAACL (2004)
Strohman, T., et al.: Indri: a language model-based search engine for complex queries. In: International Conference on Intelligence Analysis (2005)
Tur, G., De Mori, R.: Spoken Language Understanding: Systems for Extracting Semantic Information from Speech. Wiley, New York (2011)
Wang, J., Oard, D.: Matching meaning for cross-language information retrieval. Inf. Process. Manag. 48(4), 631–653 (2012)
Wegmann, S., et al.: The TAO of ATWV: probing the mysteries of keyword search performance. In: ASRU, pp. 192–197 (2013)
Weintraub, M.: Keyword-spotting using SRI’s DECIPHER large-vocabulary speech-recognition system. In: ICASSP, vol. 2, pp. 463–466 (1993)
White, R.W., Oard, D.W., Jones, G.J.F., Soergel, D., Huang, X.: Overview of the CLEF-2005 cross-language speech retrieval track. In: Peters, C., et al. (eds.) CLEF 2005. LNCS, vol. 4022, pp. 744–759. Springer, Heidelberg (2006). https://doi.org/10.1007/11878773_82
Xu, J., Weischedel, R.: Cross-lingual information retrieval using hidden Markov models. In: EMNLP, pp. 95–103 (2000)
Zbib, R., et al.: Neural-network lexical translation for cross-lingual IR from text and speech. In: SIGIR (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Nair, S., Ragni, A., Klejch, O., Galuščáková, P., Oard, D. (2020). Experiments with Cross-Language Speech Retrieval for Lower-Resource Languages. In: Wang, F., et al. Information Retrieval Technology. AIRS 2019. Lecture Notes in Computer Science(), vol 12004. Springer, Cham. https://doi.org/10.1007/978-3-030-42835-8_13
Download citation
DOI: https://doi.org/10.1007/978-3-030-42835-8_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-42834-1
Online ISBN: 978-3-030-42835-8
eBook Packages: Computer ScienceComputer Science (R0)