Classification of Seven Different Dialects Spoken in Seven Geographical Regions of Türkiye Using Machine Learning Models

Abstract

In this study, a machine learning model was developed to predict which of the seven Turkish dialects a given speech recording, collected from seven different regions of Türkiye, belongs to. The datasets used for machine learning were gathered from YouTube, prioritizing sound recordings with high potential to reflect regional dialects, focusing on natural conversations and local people’s speech. A total of 15,889 s of audio was collected, ensuring a balanced representation of each regional dialect. The features of the audio recordings, segmented into specific sizes, were extracted using MFCCs. Machine learning models were then constructed with these extracted features using 11 classification methods. With the performance enhancements obtained through optimization, the classification success for each regional dialect reached an accuracy rate of 89%, while the correct prediction rates for each of the seven regions had F1-scores ranging from 80.1% to 96.6%. The results of the analysis indicate that audio recordings from the Eastern Anatolia Region were correctly predicted at a high rate of 96.6%. This study aimed to develop a machine learning model that achieves performance improvements in identifying and predicting which regional dialects audio recordings, comprising speeches with local and regional characteristic traces in Türkiye, belong to.

Keywords

• Turkish Dialect
• Audio Data
• Machine Learning
• Classification

References

1. Abdelazim, M., Hussein, W., & Badr, N. (2022). Automatic dialect identification of spoken Arabic speech using deep neural networks. International Journal of Intelligent Computing and Information Sciences, 22(4), 25-34. https://doi.org/10.21608/ijicis.2022.152368.1207 google scholar
2. Ayvaz, U., Gürüler, H., Khan, F., Ahmed, N., Whangbo, T., & Bobomirzaevich, A. A. (2022). Automatic speaker recognition using mel-frequency cepstral coefficients through machine learning. Computers, Materials and Continua, 71(3), 5511–5521. https://doi.org/10.32604/cmc.2022.023278 google scholar
3. Brendemoen, B. (2021). Turkish dialects. In L. Johanson & E. A. Csato (Eds.), The Turkic Languages (2nd ed., 224–230). London, UK: Routledge. https://doi.org/10.4324/9781003243809-14 google scholar
4. Buran, A. (2011). Türkiye Türkçesi ağızlarının tasnifleri üzerine bir değerlendirme [An evaluation on the classification of Türkiye Turkish dialects]. Turkish Studies - International Periodical for the Languages, Literature and History of Turkish or Turkic, 6(1), 41–54. http://dx.doi.org/10.7827/TurkishStudies.1799 google scholar
5. Caucheteux, C., & King, J. R. (2022). Brains and algorithms partially converge in natural language processing. Commun. Biol., 5. https://doi.org/10.1038/S42003-022-03036-1 google scholar
6. Chien, Y. C., Wu, T. T., Lai, C. H., & Huang, Y. M. (2022). Investigation of the influence of artificial intelligence markup language-based LINE Chatbot in contextual English learning. Frontiers in Psychology, 13. https://doi.org/10.3389/FPSYG.2022.785752 google scholar
7. Das, H. C., Sarmah, K., Hajoary, D., Narzary, R., & Basumatary, R. (2023). Assamese dialect identification system using convolution neural networks. International Journal of Membrane Science and Technology, 10(2), 4340-4347. https://doi.org/10.15379/ijmst.v10i2.3519 google scholar
8. El Shazly, R. (2021). Effects of artificial intelligence on English speaking anxiety and speaking performance: A case study. Expert Systems, 38(3). https://doi.org/10.1111/EXSY.12667 google scholar

9. Ferrario, A., Luo, M., Polsinelli, A. J., Moseley, S. A., Mehl, M. R., Yordanova, K., Martin, M., & Demiray, B. (2022). Predicting working memory in healthy older adults using real-life language and social context information: A machine learning approach. Journal of Memory Research in Aging, 5(1). https://doi.org/10.2196/28333 google scholar
10. Gangarde, R., Sharma, A., & Pawar, A. (2020). Research opportunities in privacy of online social network data publishing. International Journal of Advanced Science and Technology, 29(5), 5095–5101. Retrieved from https://www.scopus.com/pages/publications/85084650314?inward= google scholar
11. García-Ordás, M. T., Alaiz-Moretón, H., Benítez-Andrades, J. A., García-Rodríguez, I., García-Olalla, O., & Benavides, C. (2021). Sentiment analysis in non-fixed length audios using a fully convolutional neural network. Biomedical Signal Processing and Control, 69. https://doi.org/10.1016/J.BSPC.2021.102946 google scholar
12. Genç, S., & Cin, A. (2016). Türkiye Türkçesi ağızlarında “Domates, kedi, patates” kelimeleri ve bir kelime haritası denemesi [An exploration of a word map: the words “Tomato, cat and potato” in the Türkiye Turkish dialects]. Mediterranean Journal of Humanities, 6(1), 185–205. https://doi.org/10.13114/MJH.2016119298 google scholar
13. Gopal, S., Bhardwaj, S., & Kiran, P. (2016). Gaussian membership function-based speaker identification using score level fusion of MFCC and GFCC. Proceedings of the International Congress on Information and Communication Technology: Advances in Intelligent Systems and Computing, 438, 283–291. https://doi.org/10.1007/978-981-10-0767-5_31 google scholar
14. Goldstein, A., Zada, Z., Buchnik, E., Schain, M., Price, A., Aubrey, B., … Hasson, U. (2022). Shared computational principles for language processing in humans and deep language models. Nature Neuroscience, 25(3), 369–380. https://doi.org/10.1038/S41593-022-01026-4 google scholar
15. Işık, G., & Artuner, H. (2018). A dataset for Turkish dialect recognition and classification with deep learning. In 2018 26th SIU,1–4. Izmir, Türkiye: IEEE. http://dx.doi.org/10.1109/SIU.2018.8404722 google scholar
16. Işık, G., & Artuner, H. (2020). Turkish dialect recognition using acoustic and phonotactic features in deep learning architectures. Journal of Information Technologies, 13(3), 207–216. https://doi.org/10.17671/gazibtd.668023 google scholar
17. Johanson, L. (2020). The classification of the Turkic languages. In M. Robbeets & A. Savelyev (Eds.), The Oxford Guide to the Transeurasian Languages, 105–114. Oxford, UK: Oxford University Press. https://doi.org/10.1093/OSO/9780198804628.003.0009 google scholar
18. Karahan, L. (2013). Gramatikal ölçütlerle belirlenen Türkiye Türkçesi ağız gruplarında leksik verilerin anlamlılığı üzerine bir araştırma [A research on the significance of lexical data in Türkiye Turkish dialect groups determined by grammatical criteria]. Diyalektolog - Ağız Araştırmaları Dergisi, 7, 1–12. Retrieved from https://dergipark.org.tr/tr/pub/diyalektolog/article/240065 google scholar
19. Karahan, L. (2022). Anadolu ağızlarının sınıflandırması [Classification of Anatolian dialects] (5th ed.). Ankara, Türkiye: Türk Dil Kurumu Yayınları. google scholar
20. Korkmaz, Y., & Boyacı, A. (2022). A comprehensive Turkish accent/dialect recognition system using acoustic perceptual formants. Applied Acoustics, 193. https://doi.org/10.1016/j.apacoust.2022.108761 google scholar
21. Koskinen, M., Kurimo, M., Gross, J., Hyvärinen, A., & Hari, R. (2020). Brain activity reflects the predictability of word sequences in listened continuous speech. NeuroImage, 219. https://doi.org/10.1016/j.neuroimage.2020.116936 google scholar
22. Lee, Y., Keating, P., & Kreiman, J. (2019). Acoustic voice variation within and between speakers. The Journal of the Acoustical Society of America, 146(3), 1568–1579. https://doi.org/10.1121/1.5125134 google scholar
23. Moondra, A., & Chahal, P. (2023). Voice feature extraction method analysis for speaker recognition with degraded human voice. Proceedings -ICAC3N 2023, 385–388. https://doi.org/10.1109/ICAC3N60023.2023.10541716 google scholar
24. Odilov, Y. (2022). Society, language and culture. Uzbekistan Language and Culture, 4(2), 6–17. https://doi.org/10.52773/tsuull.uzlc.2022.4.2/YTDE7874https://doi.org/10.52773/tsuull.uzlc.2022.4.2/YTDE7874 google scholar
25. Pednekar, M. S., & Bhattacharyya, K. (2025). Automatic identification of Malvani dialects from audio signal based on hybrid FFO-TSO with deep neural network. Multimedia Tools and Applications, 84, 17837–17869. https://doi.org/10.1007/s11042-024-19490-4 google scholar
26. Sadakata, M., & McQueen, J. M. (2013). High stimulus variability in nonnative speech learning supports formation of abstract categories: Evidence from Japanese geminates. The Journal of the Acoustical Society of America, 134(2), 1324–1335. https://doi.org/10.1121/1.4812767 google scholar
27. Scherrer, Y., van der Goot, R., & Mæhlum, P. (2025). Findings of the VarDial evaluation campaign 2025: The NorSID Shared Task on Norwegian slot, intent, and dialect identification. In Proceedings of the 12th Workshop on NLP for Similar Languages, Varieties and Dialects, 1–8. https://aclanthology.org/2025.vardial-1.1/ google scholar
28. Sharma, G., Umapathy, K., & Krishnan, S. (2020). Trends in audio signal feature extraction methods. Applied Acoustics, 158. https://doi.org/10.1016/J.APACOUST.2019.107020 google scholar
29. Soorajkumar, R., Girish, G. N., Ramteke, P. B., Joshi, S. S., & Koolagudi, S. G. (2017). Text-independent automatic accent identification system for Kannada language. Advances in Intelligent Systems and Computing, 469, 411–418. https://doi.org/10.1007/978-981-10-1678-3_40 google scholar
30. Susianti, H. W., Moon, Y. J., & Budi, I. S. (2024). The relationship between culture and language: An anthropological linguistics study. LACULTOUR: Journal of Language and Cultural Tourism, 3(2), 74–83. https://doi.org/10.52352/LACULTOUR.V3I2.1611 google scholar
31. Tanner, J. P., Takats, C., Lathan, H. S., Kwan, A., Wormer, R., Romero, D., & Jones, H. E. (2023). Approaches to research ethics in health research on YouTube: Systematic review. Journal of Medical Internet Research, 25. https://doi.org/10.2196/43060 google scholar
32. Tiwari, V., & Singhai, J. (2013). Fourier bessel and wavelet based feature extraction technique (FBWBF) for speaker identification. In C. J. Venable (Ed.), Signal Processing: New Research, 165–176. Hauppauge, NY: Nova Science Publishers. google scholar
33. Wang, Y., Spence, M. M., Jongman, A., & Sereno, J. A. (1999). Training American listeners to perceive Mandarin tones. The Journal of the Acoustical Society of America, 106(6), 3649–3658. https://doi.org/10.1121/1.428217 google scholar
34. Wei, P., He, F., Li, L., & Li, J. (2020). Research on sound classification based on SVM. Neural Computing and Applications, 32, 1593–1607. https://doi.org/10.1007/S00521-019-04182-0 google scholar
35. Zabidi, A., Mansor, W., Lee, Y. K., Mohd Yassin, A. I., & Sahak, R. (2010). Particle swarm optimization of mel-frequency cepstral coefficients computation for the classification of asphyxiated infant cry. In Proceedings of the 2010 3rd International Conference on BMEI, 3, 991–995. https://doi.org/10.1109/BMEI.2010.5639674 google scholar
36. Zhang, Q., & Hansen, J. H. L. (2018). Language/Dialect recognition based on unsupervised deep learning. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 26(5), 873–882. https://doi.org/10.1109/TASLP.2018.2797420 google scholar