Towards Fine-grained Audio Captioning with Multimodal Contextual Cues - View it on GitHub