Here is how they define them, "The surgical mask had three layers of 100% non-woven polypropylene (the exterior and interiors were spunbond and the middle layer was meltblown), elastic ear loops, and a nose bridge." What other information did you want?
But more importantly, you misunderstood why this study is so critical. Mask efficacy on population level is very hard to accurately study. Unlike a vaccine, or many studies of medications, the use of masks is not a "yes/no" situation. An N95 isn't very effective if worn below the nose, on the chin, or only used half of the day. For this reason, population studies of mask efficacy need two crucial components: 1) they need to involve a large number of participants so the variability in use does not influence the results, and 2) they need to be randomized. The study from Bangladesh meets both of these criteria - there were over 300,000 participants and mask use was randomized by village. No other study comes close to this number of participants in a randomized trial design.