The Effects Masks Have on Air Flow from Speech

22 April 2022

UC Mechanical Engineering and the Department of Linguistics demonstrate new insights into the effects of speech air flow while wearing masks.

  • SpeechAirflowCrop

The UC Department of Mechanical Engineering is delighted to announce the recent publication of Speech air flow with and without face masks, in Scientific Reports - Nature. This cross-departmental collaborative piece was produced by a team of UC Academics representing both Mechanical Engineering, the Department of Linguistics, and the New Zealand Institute of Language, Brain, and Behaviour, to seek out new insights into the effect of different commonly used face coverings (i.e., cloth masks, surgical masks, N95/K95) on speech related air flow.

Studies involving mask use over the past couple of years have largely focused on their efficacy against bacteria and viral related particle transmission, giving this team of researchers an opportunity to explore a nascent approach. By concentrating on large eddy air motion that are generated through speech, the team can distinguish between the effectiveness found using different materials of different masks. This was achieved by utilizing a single-mirror schlieren system to visually capture changes in air density produced by breathing and speech.

When discussing the efforts made between the Engineering and Linguistic Departments, Engineering Professor Mark Jermy recounted, "It's great to be working with linguists who are leading in their field. This is an application of schlieren imaging I'd never really thought about- it's great to do interdisciplinary work like this and come across new problems."

As the study related to speech production, PhD student Peiman Pishyar-Dehkordi helped account for the nuances involved with speech production by transcribing the acoustic variables associated with the study and selecting the pangram sentences that were spoken by the study’s participant. This was a crucial step to account for all the individual phones of English and provide for the “full range of turbulent and laminar air flow patterns used in English.”

Results of the study demonstrate that relatively non-porous, multi-layered mask options, such as the N95/K95 performed measurably better in preventing forward momentum of air flow compared to surgical, cloth, and especially pollen masks. However, resonating with the results found in Kellogg’s study An experimental study of the efficacy of gauze face masks from a century ago, the team demonstrated that while masks have the ability to catch and slow the spread of air flow from a speaker, they ultimately cannot stop it completely.

Senior Lecturer in the Linguistics Department and lead author Donald Derrick had this to say,  “While masks can slow down air flow, they cannot completely stop it. So, their greatest potential value is within medium risk environments…meaning a medium density of people in an indoor area, and for a short amount of time.”

The results of the study indicate that all the tested masks slowed down speech air flow when compared to not wearing a face mask. The team further suggests how “there is a trade-off in a mask’s ability to catch material (in this case air), and a tendency for material to leak around the sides of the mask.”  With these considerations in mind, mask selection may be best suited to account for an individual’s specific circumstance of potential exposure.

UC Mech Engineering would like to thank the team for their recent contribution and anticipates  hearing about any further development from this study in the future.

Article Link


Mark Jermy Covid Article Model

UC Engineering Research Highlighted in COVID-19 Government Report

The UC Mech Engineering Department recognizes the contributions made by staff and students in NZ COVID-19 report

Aerospace Course Jet Photo

Design and analysis of a long-range business jet 

The first group of University of Canterbury (UC) students minoring in Aerospace Engineering got off to a flying start with the design of new ...