There are two types of airflow important for the production of speech sounds.

  1. Laminar flow where the air particles all move in the same direction and in essentially parallel lines. This sort of airflow is found in sonorant sounds such as vowels, approximants and nasals.
  2. Turbulent flow where the movement of the air particles becomes chaotic and eddies are formed.  This sort of flow is found in obstruent sounds such as fricatives, the fricative portions of affricates and the burst noise of plosives.

A quantity called the Reynold’s Number (Re) is a fairly good predictor of whether the flow in a pipe (such as the vocal tract) will be laminar or turbulent.  The relevant equation for the number is:

Re = QD/νA

where Q is the volumetric flow rate in m3/s, D is the diameter of the pipe in m, v is the kinematic viscosity of air in m2/s, and A is the cross-sectional area of the pipe in m2.

It is easy to see that an increase in Q and/or a decrease in A will cause an increase in Re. When Re < 2300 laminar flow usually results. Turbulent flow is usual when Re > 4000. For intermediate values of Re both laminar and turbulent flow is possible, depending on factors like surface roughness and flow uniformity.