Muhammad Hafiz, Radzali (2023) A STUDY ON HYPERPARAMETER TUNING FOR TEACHABLE MACHINE ON AN ISOLATED WORD SPEECH RECOGNITION FOR CHILDREN’S SPEECH. [Final Year Project Report] (Unpublished)
PDF (Please get the password by email to repository@unimas.my , or call ext: 082-583914/3973/3933)
Muhammad_Hafiz ft.pdf Restricted to Registered users only Download (2MB) |
Abstract
Automatic speech recognition is a widely used technology that is currently being used for many purposes such as accessibility for visually impaired individuals, ease of access towards systems and many more. This paper is to show and discuss the findings of the Final Year Project study of using a Teachable Machine for researching the patterns of native and non-native English speakers additionally the effects of manipulation of Hyperparameter Tuning towards the results of the Teachable Machine audio model. The non-native English speaker’s dataset is based on the audio of non-native English children from Sarawak, Malaysia aged 7 to 12 years old reading a set list of isolated English words. The results from the Teachable Machine are shown and the production of the ReactJS web-based system is being made with the conclusion of the project being successful but does require more training of datasets
Item Type: | Final Year Project Report |
---|---|
Additional Information: | Project report (B.Sc.) -- Universiti Malaysia Sarawak, 2023. |
Uncontrolled Keywords: | Automatic Speech Recognition, Hyperparameter Tuning, Isolated Word Speech, Machine Learning, Teachable Machine |
Subjects: | Q Science > QA Mathematics > QA76 Computer software |
Divisions: | Academic Faculties, Institutes and Centres > Faculty of Computer Science and Information Technology Faculties, Institutes, Centres > Faculty of Computer Science and Information Technology Academic Faculties, Institutes and Centres > Faculty of Computer Science and Information Technology |
Depositing User: | Unai |
Date Deposited: | 17 Jan 2024 06:18 |
Last Modified: | 17 Jan 2024 06:18 |
URI: | http://ir.unimas.my/id/eprint/44170 |
Actions (For repository members only: login required)
View Item |