Voice Activity Detection and Keyword Classification Using Data from the Intraoral Densor Sensing Platform

Using Hidden Markov Models to Detect Speech Activity and Recognize Keywords in Intraoral Sensor Data

Bachelor Thesis (2025)
Author(s)

M.J.N. Klumpenaar (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Contributor(s)

Vivian Dsouza – Mentor (TU Delft - Embedded Systems)

P Pawełczak – Mentor (TU Delft - Embedded Systems)

J.A. Pouwelse – Graduation committee member (TU Delft - Data-Intensive Systems)

Faculty
Electrical Engineering, Mathematics and Computer Science
More Info
expand_more
Publication Year
2025
Language
English
Graduation Date
01-07-2025
Awarding Institution
Delft University of Technology
Project
['CSE3000 Research Project']
Programme
['Computer Science and Engineering']
Faculty
Electrical Engineering, Mathematics and Computer Science
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

The Densor is an intraoral sensor platform created to capture unique data from inside the human mouth. This thesis studies the possibility of using Densor-recorded sensor data for Voice Activity Detection (VAD) and basic non-acoustic speech (keyword) recognition. It is part of a broader effort to explore the different use cases of the Densor. This thesis summarizes the performance of existing non-acoustic wearable speech recognition devices, describes the available Densor data, details feature extraction and selection, and shows how Hidden Markov Models can beusedfor both VADandkeyword recognition. The results are promising, with an F1-Score of up to 0.73 for VAD and up to 75% accuracy for keyword recognition. While these results show that both tasks are possible, they cannot be generalized due to the small and unvaried dataset. Future work suggestions include expanding and increasing the variety of the dataset and exploring alternative models such as Conditional Random Fields and Recurrent Neural Networks.

Files

License info not available
warning

File under embargo until 22-06-2027