Abstract
This article is dedicated to a new method for translating audio recordings of interjections between two domains. The original speaker's voice is transformed to the target one within a given voice pair, and vice versa. Basing on an overview conducted, mel-frequency cepstral coefficients are selected as main features of signal. The introduced method is implemented and approbated on the grounds of a book-reading training dataset and sample interjections for a number of voice pairs, including human-robotic. Recommendations are given on the method applicability and on dataset recording. Obtained results testify that we found the solution to the problem of overcoming constraints of existing speech-synthesizing software, namely that of the limitedness of interjections forms and that of the poor intonations variety. The method is to be applied in order to fill reactions databases for dialogue systems designed for affective communication, such as the F-2 interlocutor robot. The introduced method will enable spoken reactions corpora developers to record interjections of interest and to translate them to the selected synthetic voice.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ahirwar, K.: Generative Adversarial Networks Projects. Packt Publishing, Birmingham (2019)
Antsiferova, V.I., Pesetskaya, T.V., Yuldoshev, I.I., Syanyan, L., Tsin, V., Lavlinskiy, V.V.: One of approaches towards determining relevant features for audiosignals: a sample research on interjections (in Russian). In: Zol’nikov, V.K. (ed.) Sovremennye aspekty modelirovaniya sistem i protsessov: Materialy Vserossiyskoy nauchno-prakticheskoy konferentsii, 2021, pp. 15–20. Voronezh State University of Forestry and Technologies named after G.F. Morozov, Voronezh (2021)
Becker, C.W., Kopp, S., Wachsmuth I.: Simulating the emotion dynamics of a multimodal conversational agent. In: Affective Dialogue Systems, Tutorial and Research Workshop, ADS 2004, LNCS 3068, pp. 154–165. Springer, Heidelberg (2004)
Bloomfield, L.: An introduction to the study of language. Holt, New York (1914)
Cai, Y.: Empathic Computing. In: Cai, Y., Abascal, J. (eds.) Ambient Intelligence in Everyday Life. LNCS (LNAI), vol. 3864, pp. 67–85. Springer, Heidelberg (2006). https://doi.org/10.1007/11825890_3
Campbell, N.: Recording techniques for capturing natural everyday speech. In: Proc. Language Resources and Evaluation Conference (LREC-02), pp. 2029–2032. European Language Resources Association, Paris (2002)
Campbell, N.: Extra-semantic protocols; input requirements for the synthesis of dialogue speech. In: Proc. Affective Dialogue Systems, Tutorial and Research Workshop, ADS 2004, LNCS 3068, pp. 221–228. Springer, Heidelberg (2004)
Dauphin, Y.N., Fan, A., Auli, M., Grangie, D.: Language modeling with gated convolutional networks. In: Precup, D., Teh, Y.W. (eds.) Proc. of the 34th International Conference on Machine Learning, PMLR 70. MLResearchPress (2017)
Davis, S.B., Mermelstein, P.: Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences. IEEE Trans. Acoust. Speech Signal Process. 28(4), 357–366 (1980)
Dingemanse, M.: Interjections (preprint). In: van Lier, E. (ed.) The Oxford Handbook of Word Classes. Oxford University Press, Oxford (2021)
Drijvers, L., Holler, J.: The multimodal facilitation effect in human communication. Psychon. Bull. Rev. 30(2), 792–801 (2022)
Efimov, A.P., Nikonov, A.V., Sapozhkov, M.A., Shorov, V.I.: Acoustics: Spravochnik (in Russian). In: Sapozhkov, M.A. (ed.) Radio i svyaz’, Moskva (1989)
Elffers, E.: Interjections and the language functions debate. Asia Paci. J. Human Resou. 50(1), 17–29 (2008)
Goffman, E.: Response cries. Language 54(4), 787–815 (1978)
Goodfellow, I., et al.: Generative adversarial networks. Advances in Neural Information Processing Systems (NIPS 2014) 27, 2672–2680 (2014)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778. IEEE, Piscataway (2016)
Huang, X., Acero, A., Hon, H.-W.: Spoken language processing: a guide to theory, algorithm, and system development. Prentice Hall PTR, New Jersey (2001)
Ippolitova, N.A.: Ritorics. Prospekt, Moscow (2013). (in Russian)
Isola, P., Zhu, J.-Y., Zhou, T., Efros, A.A.: Image-to-image translation with conditional adversarial networks. In: CVPR 2017, pp. 5967–5976. IEEE, Piscataway (2017)
Kollias, D., Zafeiriou, S.: Expression, affect, action unit recognition: Aff-wild2, multi-task learning and arcface. In: 30th British Machine Vision Conference, pp. 78.1–78.15. BMVA Press, Durham (2019)
Kozlov, A.V., Kudashev, O., Matveev, Y.: A system for dictors identification by voice for NIST SRE 2012 contest (in Russian). Trudy SPIIRAN 2, 350–370 (2013)
Kudashev, O.Y.: A system for dictors separation based on probabilistic linear discriminant analysis. Ph.d. thesis (in Russian). ITMO, Saint-Petersburg (2014)
Liu, K., Zhang, J., Yan, Y.: High quality voice conversion through phoneme-based linear mapping functions with straight for mandarin. In: Fourth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD 2007), vol. 4, pp. 410–414. IEEE, Piscataway (2007)
Malkina, M., Zinina, A., Arinkin, N., Kotov, A.: Multimodal hedges for companion robots: a politeness strategy or an emotional expression? In: Selegey, V.P., et al. (eds.) Computational Linguistics and Intellectual Technologies: Papers from the Annual International Conference “Dialogue”, issue 22, pp. 319–326. RSUH, Moscow (2023)
Morise, M., Yokomori, F., Ozawa, K.: WORLD: a vocoder-based high-quality speech synthesis system for real-time applications. IEICE transactions on information and systems, E99-D(7), 1877–1884 (2016)
Morise, M.: Platinum: A method to extract excitation signals for voice synthesis system. Acoustic Science & Technology 33(2), 123–125 (2012)
Morise, M.: Cheaptrick, a spectral envelope estimator for high-quality speech synthesis. Speech Commun. 67, 1–7 (2015)
Morise, M.: Harvest: a high-performance fundamental frequency estimator from speech signals. In: Proc. INTERSPEECH 2017, pp. 2321–2325. ISCA, Baixas (2017)
Morise, M.: Implementation of sequential real-time waveform generator for high-quality vocoder. In: Proceedings of APSIPA Annual Summit and Conference, pp. 821–825. IEEE, Pictasaway (2020)
Niewiadomski, R., Bevacqua, E., Mancini, M, Pelachaud, C.: Greta: an interactive expressive ECA system. In: Proceedings of The 8th International Conference on Autonomous Agents and Multiagent Systems, vol. 2, pp. 1399–1400. International Foundation for Autonomous Agents and Multiagent Systems, Richland (2009)
Ohtani, Y., Toda, T., Saruwatari, H., Shikano, K.: Maximum likelihood voice conversion based on GMM with STRAIGHT mixed excitation. In: Proc. INTERSPEECH 2006, pp. 2266–2269. IEEE, Pictasaway (2007)
Plotnikov, V.N., Sukhanov, V.A., Zhigulevtsev, Y.: Spoken dialogue in control systems. Mashinostroeniye, Moscow (1988). (in Russian)
Ronzhin, A.L., Karpov, A.A., Lee, I.V.: Speech and multimodal interfaces. Nauka, Moscow (2006). (in Russian)
RusCorpora: the Russian National Corpus of texts, https://ruscorpora.ru/en/, last accessed 15 June 2023
Sahidullah, M., Saha, G.: Design, analysis and experimental evaluation of block based transformation in MFCC computation for speaker recognition. Speech Commun. 54(4), 543–565 (2012)
Shröder, M.: Towards a standards-based framework for building emotion-oriented systems. Advances in Human-Computer Interaction 2010, article ID 319406 (2010)
Stijovic, R., Lazic-Konjik, I., Spasojevic, M.: Interjections in the contemporary Serbian language: classification and lexicographic treatment (in Serbian). Juznoslovenski filolog 75, 37–61 (2019)
Volkova, L.S., Shakhovskaya, S.N. (eds.): Logopediya: Uchebnik dlya studentov defektologicheskikh fakul’tetov (Fakul’tetov korrektsionnoj pedagogiki) pedagogicheskikh universitetov i institutov, 3rd edn. Gumanit. izd. tsentr VLADOS, Moscow (1998). (in Russian)
Volkova, L., Ignatev, A., Kotov, N., Kotov, A.: New communicative strategies for the affective robot: F-2 going tactile and complimenting. In: Creativity in Intelligent Technologies and Data Science, CCIS 1448, pp. 163–176. Springer, Heidelberg (2021)
Wharton, T.: Pragmatics and non-verbal Communication. Cambridge University Press, Oxford (2009)
Wierzbicka, A.: The semantics of interjection. J. Pragmat. 18(2–3), 159–192 (1992)
Yandex LLC: Yandex Alice, https://yandex.ru/dev/dialogs/alice/doc/nlu.html, last accessed 16 March 2023
Yandex LLC: Yandex SpeechKit, https://cloud.yandex.ru/services/speechkit, last accessed 16 March 2023
Zhu, J.-Y.: How to know the training should stop. CycleGAN and pix2pix in PyTorch/Jun-Yan Zhu github, https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/issues/166, last accessed 16 March 2023
Zhu, J.-Y., Park, T., Isola, P., Efros, A.A.: Unpaired image-to-image translation using cycle-consistent adversarial networks. In: 2017 IEEE International Conference on Computer Vision (ICCV), pp. 2242–2251. IEEE, Piscataway (2017)
Acknowledgements
This research is supported by the grant of the Russian Science Foundation № 19-18-00547, https://rscf.ru/project/19-18-00547/.
We would like to thank our volunteer speakers who helped us recording a dataset, and our assessors whose opinions are priceless to our sort of projects, when it comes to crowdsourcing (only the final CycleGan results assessments are stated in this article; inbetween translations were estimated thoroughly as well). We also express gratitude to Edward Klyshinsky for his repeated affirmative “mhm”, which led us to the answer to the puzzling “oohoo” question.
We would like to thank Alexey Yuryevich Popov from IU-6 department, BMSTU, for his kind permission to use his devices of high computational capability in order to train our CycleGANs, and Kirill Leonidovich Tassov from IU-7 department, BMSTU, for his priceless remarks, which helped us to improve our results.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Polianskaya, L., Volkova, L. (2023). A CycleGAN-Based Method for Translating Recordings of Interjections. In: Kravets, A.G., Shcherbakov, M.V., Groumpos, P.P. (eds) Creativity in Intelligent Technologies and Data Science. CIT&DS 2023. Communications in Computer and Information Science, vol 1909. Springer, Cham. https://doi.org/10.1007/978-3-031-44615-3_6
Download citation
DOI: https://doi.org/10.1007/978-3-031-44615-3_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44614-6
Online ISBN: 978-3-031-44615-3
eBook Packages: Computer ScienceComputer Science (R0)