Approximants are speech sounds which are pronounced with a stricture of open approximation. This means that the active articulator approaches the passive articulator but there remains a wide gap between both so that the airflow does not cause an audible hiss. Approximants can have a central or a lateral airflow. In central approximants, the air flows along a central path over the tongue. In lateral approximants the tongue touches the roof of the mouth in the central area and the air flows around this obstacle so that the airflow is lateral.
 
The IPA distinguishes 5 symbols for the central approximants at the following places of articulation: labio-dental, alveolar, retroflex, palatal and velar. In addition, it provides 4 symbols for lateral approximants: alveolar, retroflex, palatal and velar. Additional symbols for the central approximants can be derived from the fricative symbol at the corresponding place of articulation by adding a diacritic denoting a more open degree of approximation: this can be seen above for e.g. the lingo-labial and dental approximants.

These are the five most frequent approximants in the languages of the world:

                           [ j ] > [ w ] > [ l ] > [ ɭ ] > [ ɻ ]

Approximants are typically voiced, but voiceless approximants do occur. Iaai, which is a language spoken in New Caldedonia in the South Pacific, reportedly has several voiceless approximants which contrast phonemically with voiced approximants (Maddieson & Anderson, 1994. Phonetic Structures of Iaai. UCLA Working Papers in Phonetics, 87:163-182).

The videos below illustrate a few approximants in languages like Dutch and English. Notice in particular for [j] and [w] the relatively wide space between the highest point of the tongue and the roof of the mouth. This results in very little obstruction of the airstream so that there is no audible frication to these sounds. Furthermore, the velum is raised and the vocal folds vibrate in all these examples.