Fully Quantum Vowel Recognition
DOI:
https://doi.org/10.24425/ijet.2026.6339Abstract
This paper presents the results of using a quantum machine learning algorithm for the task of vowel recognition. It compares computation times and accuracy results of different circuit depths in a simulated quantum environment. The paper proposes a fully quantum neural network using 4 qubits as feature maps with formant frequencies as input features. It compares the results of binary classification and the more problematic multiclass classification in quantum machine learning. The results show very high accuracy of around 97% for binary, 80% for 3 classes, 70% for 4 classes, and 50% for all 5 classes. The quantum neural networks were first pre-trained using a simulator and then evaluated on the real quantum computer Odra 5 to assess the impact of noise and hardware configuration on the performance of the trained quantum circuits.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 International Journal of Electronics and Telecommunications

This work is licensed under a Creative Commons Attribution 4.0 International License.