Method and system for laughter detection

A method for detecting an at least one laughter episode in an audio signal of a telephone interaction in a call center, the method comprising the steps of:

• capturing an audio signal of a telephone interaction in a call center;
• detecting, an at least one burst in the audio signal;
• detecting, at least predetermined number of consecutive bursts as N-tuple;
• extracting in time domain an at least one time domain feature from the at least predetermined number of consecutive bursts, wherein the at least one time domain feature is selected from the group consisting of:
• mean duty cycle, mean distance between voiced sections, and an itakura distance;
• scoring the least one time domain feature against an at least one predetermined threshold;
• determining whether the at least one feature complies with the at least one predetermined threshold;
• and
• if the at least one time domain feature complies with the at least one predetermined threshold, determining whether the at least predetermined number of consecutive bursts represent a laughter episode.