Single cane reeds used for playing clarinets or saxophones are described by makers by their strength and their shapes (cutting). For reeds assumed to be identical according to the maker (same strength, same shape), strong differences in the perceived quality is expressed by musicians. In this context, the experimental characterisation of reeds (from the perceptive and objective sides) is a key issue for reed makers in order to better predict reed musical abilities. This paper present and discuss different measurement methods. These methods can be divided into three families : - measurement of the reed alone through static or dynamic measurements in order to derive stiffness and modal parameters - measurement of the "embouchure" (reed+mouthpiece+lip) alone through static or quasi-static measurements in order to estimate the non linear characteristics, the non linear reed stiffness or other mechanical parameters, - measurements with the complete instrument artificially played in order to derive either the bifurcation diagrams or some playing parameters (frequency, spectral centroid, pressure thresholds). The results obtained with the different methods are compared and the correlation with subjective descriptors is given. Finally it appears that the characterisation of the "embouchure" is required and can be done in a static manner.
