Skip to main content

An ASR Corpus in Chhattisgarhi, a Low Resource Indian Language

  • Conference paper
  • First Online:
Speech and Computer (SPECOM 2023)

Abstract

RESPIN is a project that aims at the development of a dialect-rich database and some user-friendly voice-technology applications in 9 Indian languages including Chhattisgarhi. The paper elaborates on the entire process of such a low-resource database preparation in a crowd-sourced manner . Through this work we have open-sourced around 250 h of dialect-rich, domain-rich Chhattisgarhi ASR dataset to popularize the scope of voice technology to the Chhattisgarh population. The paper also describes the development of a base model with a WER score of 11.58% on the test set.

Supported by Bill & Melinda Gates Foundation (BMGF).

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 69.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 89.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    https://respin.iisc.ac.in/.

  2. 2.

    https://data.ldcil.org/speech.

  3. 3.

    https://github.com/kaldi-asr/kaldi/blob/master/egs/wsj/s5/local/chain2/tuning/run_tdnn_1i.sh.

References

  1. Gumperz, J.J.: Speech variation and the study of Indian civilization. Am. Anthropol. 63(5), 976–988 (1961)

    Article  Google Scholar 

  2. Londhe, N.D., Kshirsagar, G.B.: Speaker independent isolated words recognition system for chhattisgarhi dialect. In: 2017 International Conference on Innovations in Information, Embedded and Communication Systems (ICIIECS), pp. 1–6 (2017). https://doi.org/10.1109/ICIIECS.2017.8276169

  3. Londhe, N.D., Kshirsagar, G.B., Tekchandani, H.: Deep convolution neural network based speech recognition for chhattisgarhi. In: 2018 5th International Conference on Signal Processing and Integrated Networks (SPIN), pp. 667–671 (2018). https://doi.org/10.1109/SPIN.2018.8474064

  4. Londhe, N., Kshirsagar, G.: Chhattisgarhi speech corpus for research and development in automatic speech recognition. Int. J. Speech Technol. 21, 193–210 (2018)

    Article  Google Scholar 

  5. Mandal, R.B.: Patterns of regional geography: an international perspective \(\cdot \), vol. 2, p. 140 (1990)

    Google Scholar 

  6. Maps of India.com: Economy of chhattisgarh. https://www.mapsofindia.com/chhattisgarh/economy/, (Accessed 31 July 2023)

  7. MP Tour and Travels.com: Languages of chhattisgarh, chhattisgarh tourism. https://www.chhattisgarhtourism.co.in/languages-of-chhattisgarh.html, (Accessed 31 July 2023)

  8. Office of the registrar general: Census of India 2011. https://censusindia.gov.in/2011Census/C-16_25062018_NEW.pdf, (Accessed 07 October 2021)

  9. Povey, D., et al.: The Kaldi Speech recognition toolkit. In: Proceedings of ASRU (December 2011)

    Google Scholar 

  10. Povey, D., et al.: Purely Sequence-Trained Neural Networks for ASR Based on Lattice-Free MMI. In: Proceedings of Interspeech 2016, pp. 2751–2755 (2016). https://doi.org/10.21437/Interspeech.2016–595

  11. Sethi, N., Dev, A.: Survey on automatic speech recognition systems for indic languages. In: Dev, A., Agrawal, S.S., Sharma, A. (eds.) AIST 2021. CCIS, vol. 1546, pp. 85–98. Springer, Cham (2022). https://doi.org/10.1007/978-3-030-95711-7_8

    Chapter  Google Scholar 

  12. Stolcke, A.: SRILM - An extensible language modeling toolkit. In: INTERSPEECH (2002)

    Google Scholar 

  13. Vydana, H.K.: Salient Features for Multilingual Speech Recognition in Indian Scenario. Ph.D. thesis (Jan 2019). http://hdl.handle.net/10603/246056

  14. World Bank staff estimates based on the United Nations Population Divisions, World Urbanization Prospects: Rural population (\(\%\) of total population) - india. https://data.worldbank.org/indicator/SP.RUR.TOTL.ZS?locations=IN, (Accessed 31 July 2023)

Download references

Acknowledgements

We thank everyone who supported us throughout this study. We especially thank the funding agency BMGF, the validators, and other volunteers who contributed to collecting the database.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to G. Deekshitha .

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Singh, A. et al. (2023). An ASR Corpus in Chhattisgarhi, a Low Resource Indian Language. In: Karpov, A., Samudravijaya, K., Deepak, K.T., Hegde, R.M., Agrawal, S.S., Prasanna, S.R.M. (eds) Speech and Computer. SPECOM 2023. Lecture Notes in Computer Science(), vol 14339. Springer, Cham. https://doi.org/10.1007/978-3-031-48312-7_14

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-48312-7_14

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-48311-0

  • Online ISBN: 978-3-031-48312-7

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics