Abstract

Due to prevalent vocal health issues in teachers, the acoustics of K-12 classrooms has become a topic of study in acoustics. One way to understand the effects of a classroom's physical space on speech is with vocal effort studies. This thesis aims to enable these studies, without the need to move a talker from room to room, by creating auralizations through real-time convolution of speech with oral binaural room impulse responses (OBRIRs). These auralizations can be used to test talkers inside an anechoic chamber as they experience speaking in different acoustical environments. A system that can successfully execute convolution in real time requires finely-tuned parameters and an optimized algorithm. Efforts and lessons learned during the development of this system are shared. Finally, results from preliminary testing of talkers located inside classrooms are shared (the goal was to compare these results to those obtained using real-time convolution system (RTCS) simulations using these same OBRIRs); the data from these in-classroom tests provides lessons learned that can inform future vocal strain tests, those made in classrooms and with an RTCS, to ensure less variability and clearer trends in the results.

Degree

MS

College and Department

Physics and Astronomy; Computational, Mathematical, and Physical Sciences

Rights

https://lib.byu.edu/about/copyright/

Date Submitted

2024-12-20

Document Type

Thesis

Keywords

real-time convolution, auralization, classroom acoustics, oral binaural room impulse response, STM32F746 discovery board, ping-pong buffer, block convolution, overlap-save method, non-uniform partitioned convolution, filter partitioning, vocal strain, Lombard speech, jitter, shimmer

Language

english

Share

COinS