Emotion and theme recognition is a popular task in music information retrieval that is relevant for music search and recommendation systems. In the frame of MediaEval workshop, the MTG proposes the challenge Emotion and Themes in Music, that invites the participants to try their skills at recognizing moods and themes conveyed by the audio tracks. This task involves the prediction of moods and themes conveyed by a music track, given the raw audio. The examples of moods and themes are: happy, dark, epic, melodic, love, film, space etc. Each track is tagged with at least one tag that serves as a ground-truth.

Participants are expected to train a model that takes raw audio as an input and outputs the predicted tags. To solve the task, participants can use any audio input representation they desire, be it traditional handcrafted audio features or spectrograms or raw audio inputs for deep learning approaches. We also provide a handcrafted feature set extracted by the Essentia audio analysis library as a reference. We allow usage of third-party datsets for model development and training, but it needs to be mentioned explicitly.

The task targets researchers in music information retrieval, music psychology, machine learning, and music and technology enthusiasts in general.

The task organizers are Philip Tovstogan, Dmitry Bogdanov, Alastair Porter

MediaEval is a benchmarking initiative dedicated to evaluating new algorithms for multimedia analysis, retrieval and exploration. It emphasizes the ‘multi’ in multimedia (by involving multiple modalities such as visual, textual, audio, and sensor data) and focuses on human and social aspects of multimedia tasks. The MediaEval 2021 Workshop will be held completely online this year and participation is free. However, you are required to register in order to attend.

Information: https://multimediaeval.github.io/editions/2021/