I'm looking for labelled audio data. Like a cat meowing, or a spoon falling on the ground, a car driving past, etc... i.e. sound clips of events. Does anyone know of where to find this? Would be very grateful for any help here.
As with many new areas of research, this doesn't seem to be a domain with much truly open data, but you may be able to contact academic researchers about using datasets they have compiled for their work.
This W3C wiki page on sentiment analysis lists several such datasets.
For example, the Center for the Study of Emotion and Attention at University of Florida describes the International Affective Digital Sound (IADS) system:
The International Affective Digitized Sound system (IADS) provides a set of acoustic emotional stimuli for experimental investigations of emotion and attention. This set of standardized, emotionally-evocative, internationally accessible sound stimuli includes contents across a wide range of semantic categories. The IADS (pronounced "eye-ads") is being developed and distributed by the Center for Emotion and Attention (CSEA) at the University of Florida.
See their page for details on requesting the data set.
This Kaggle competition refers to several video datasets, but usable in analyzing the audio component. One of them is MOSI: Multimodal Corpus of Sentiment Intensity and Subjectivity Analysis in Online Opinion Videos I don't see an obvious channel for requesting that data, but you could reach out to the authors of the paper described at that link.