A Comparative Analysis of Recurrent Neural Networks-LSTM and 1D Convolutional Neural Network in Wake Word Detection System of Regional Dialects


  • Chaitra G. P., Shylaja S. S.


MFCC, CONV 1D, Dialect Identification, RNN(LSTM), Trainable Parameters, WWD.


This paper aims in comparative analysis of two deep neural network techniques -RNNs and 1D-CNN models  in building the WWD system in Kannada for five various locations with their dialects in the state of Karnataka and they are:- Dharwad, Tulu, Dogg anal, Urban Kannada in addition the Kodagu region as well. The customized WWD system is built using Conv1d model with 97% accuracy compared to RNNs with 42.5% precision. The variation present with local dialects are finely specified by 1d CNN in analogous study along with RNN model in verifying on the dialect dataset on the labels impending which are contrasting and the implementation of Conv1d makes better predictions on the Idiom Dataset.


Download data is not yet available.


H. C. Das and U. Bhattacharjee, "Assamese Dialect Identification using," in IEEE World Conference on Applied Intelligence and Computing (AIC, 2022.

Kumar, Rajath et al. “On Convolutional LSTM Modeling for Joint Wake-Word Detection and Text Dependent Speaker Verification.” Interspeech (2018)..

H. Wang, M. Cheng, Q. Fu and M. Li, "The Dku Post-Challenge Audio-Visual Wake Word Spotting System," arXiv, 4 March 2023.

Y. Tian, H. Yao, M. Cai, Y. Liu and Z. Ma, "Improving RNN Transducer Modeling for Small-Footprint Keyword Spotting," ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Toronto, ON, Canada, 2021, pp. 5624-5628.

J. Lee, K. Kim and M. Chung, "Korean Dialect Identification Based on Intonation Modeling," in 24th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA),, Singapore, 2021.

Lokitha, Iswarya, Archana and A. Kumar, "Smart Voice Assistance for Speech disabled and Paralyzed People," in International Conference on Computer Communication and Informatics (ICCCI ), Coimbatore, 2022.

Y. Wang, H. Lv, D. Povey, L. Xie and S. Khudanpur, "Wake Word Detection with Alignment-Free Lattice-Free MMI," INTERSPEECH 2020, 25-29 October 2020.

T.-H. Tsai and P.-C. Hao, "Customized Wake-Up Word with Key Word Spotting using Convolutional Neural Network," in IEEE, 2019.

S. Sarkar, B. Kumar and S. Kumar, "Mobile Applications for Indian Agriculture and Allied Sector:An Extended Arm for Farmers," International Journal of Current Microbiology and Applied Sciences, vol. 10, no. 3, 2021.

M. Tzudir, S. Baghel, P. Sarmah and S. R. M. Prasanna, "Analyzing RMFCC Feature for Dialect Identification in Ao, an Under-Resourced Language," 2022.

N. C. Diaz, N. Sasaki, T. W, Tsusaka and S. Szabo, "Factors affecting farmers’ willingness to adopt a mobile app in the marketing of bamboo products," Science Direct, vol. 11, 2021.

R. K. Raman, D. K. Singh, U. Kumar and S. Sarkar, "Agricultural Mobile Apps for Transformation of Indian Farming," ReserachGate, vol. 07, no. 04, April 2021.

S. G. Mane and K. R.V, "Design and Development of Mobile App for Farmers," International Journal of Trend in Scientific Research and Development (IJTSRD), pp. 179-182, 2019.

R. Kumar, "Farmers’ Use of the Mobile Phone for Accessing Agricultural Information in Haryana: An Analytical Study," Open Information Science, 7 April 2023.

K. D. M, and S. K. R. M, "FARMER’S ASSISTANT using AI Voice Bot," 2021 3rd International Conference on Signal Processing and Communication (ICPSC), pp. 527-531, 2021.

Z. Dan, Y. Zhao, X. Bi and Q. Ji, "Multi-Task Transformer with Adaptive Cross-Entropy Loss for Multi-Dialect Speech Recognition," MDPI, 8 OCTOBER 2022.

R. Z. Qiuchen Yu, "Wake Word Detection Model Based on Res2Net," JOURNAL OF LATEX CLASS FILES, vol. 10, no. 10, 30 September 2022.

Y. Wang, H. Lv, D. Povey, L. X. and S. Khudanpur, "WAKE WORD DETECTION WITH STREAMING TRANSFORMERS," in IEEE, Toronto, Canada, 2021.

D. Landmann, C. Lagerkvist and V. Otter, "Determinants of Small Scale Farmers’ Intention to Use Smartphones for Generating Agricultural Knowledge in Developing Countries: Evidence from Rural India," The European Journal of Development Research, 10 August 2020.

M.L.Dhore and M. Dhakate, "Insurance Value Chain Chatbot for Farmers," in ResearchGate, 2022.

C. Li, L. Zhu, S. Xu, P. Gao and B. Xu, "Recurrent Neural Network Based Small-footprint Wake-up-word Speech Recognition System with a Score Calibration Method," 2018 24th International Conference on Pattern Recognition (ICPR), Beijing, China, 2018, pp. 3222-3227.

V. Ribeiro, Y. Huang, Y. Shangguan, Z. Yang, L. Wan and M. Sun, "Handling the Alignment for Wake Word Detection:A Comparison Between Alignment-Based, Alignment-Free and Hybrid Approaches," in Accepted to Interspeech 2023, 2023.

C. R. Kinkar and Y. K. Jain, "AN OVERVIEW OF MODERN ERA SPEECH RECOGNITION MODEL," International Journal of Creative Research Thoughts (IJCRT), vol. 9, no. 9, September 2021.

T.Cynthia and C. Newton, "Voice Based Answering Technique for Farmers in Mobile Cloud Computing," International Journal of Scientific Research in Computer Science Applications and Management Studies, vol. 7, no. 3, 13 JULY 2020.

D. Rostami and Y. Shekofteh, "A Persian Wake Word Detection System Based on the Fine Tuning of A Universal Phone Decoder and Levenshtein Distance," 2023 9th International Conference on Web Research (ICWR), Tehran, Iran, Islamic Republic of, 2023, pp. 35-40.

M. S. R. M. C. P. P. D. N. Arnav Kundu, "HEiMDaL: Highly Efficient Method for Detection and Localization of wake-words," in Audio and Speech Processing-ICASSP, 2023.




How to Cite

Chaitra G. P. (2024). A Comparative Analysis of Recurrent Neural Networks-LSTM and 1D Convolutional Neural Network in Wake Word Detection System of Regional Dialects. International Journal of Intelligent Systems and Applications in Engineering, 12(21s), 3742 –. Retrieved from https://ijisae.org/index.php/IJISAE/article/view/6124



Research Article