v2.0

Published updated detection model that is more accurate and more robust to false positives as a result of background noise. Also release a set of annotations for evaluating laughter detection. Updates are described in the Interspeech 2021 paper.