One embodiment of the present invention sets forth a technique for modifying an audio parameter based on a gesture. The technique includes acquiring sensor data associated with a hand of a user and analyzing the sensor data to determine at least one hand position. The technique further includes detecting a hand gesture based on the at least one hand position and, in response to the hand gesture, modifying a spatial audio parameter associated with an audio stream to generate a modified audio stream. The technique further includes causing the modified audio stream to be reproduced for output to the user.