Happy or Evil Laughter? Analysing a Database of Natural Audio Samples

TL;DR

This study analyzes a curated laughter audio dataset to distinguish positive from negative laughter using phonetic analysis and machine learning, achieving up to 70% recall in classification.

Contribution

It introduces a new annotated laughter dataset and compares phonetic and machine learning approaches for emotion classification in natural audio samples.

Findings

Best models achieve 70% unweighted average recall

Phonetic features provide insights into laughter types

Machine learning effectively classifies positive and negative laughter

Abstract

We conducted a data collection on the basis of the Google AudioSet database by selecting a subset of the samples annotated with \textit{laughter}. The selection criterion was to be present a communicative act with clear connotation of being either positive (laughing with) or negative (being laughed at). On the basis of this annotated data, we performed two experiments: on the one hand, we manually extract and analyze phonetic features. On the other hand, we conduct several machine learning experiments by systematically combining several automatically extracted acoustic feature sets with machine learning algorithms. This shows that the best performing models can achieve and unweighted average recall of .7.