Learning Contextual Tag Embeddings for Cross-Modal Alignment of Audio and Tags
This work addresses cross-modal alignment for audio and tags, which is incremental as it builds on existing self-supervised methods by incorporating attention mechanisms for better generalization.
The paper tackled the problem of learning audio representations that align with tags, including unknown ones, by proposing a method using an audio autoencoder, word embeddings, and multi-head self-attention with contrastive loss. The result showed that this approach improved audio representations, as evidenced by performance in sound, music genre, and music instrument classification tasks.
Self-supervised audio representation learning offers an attractive alternative for obtaining generic audio embeddings, capable to be employed into various downstream tasks. Published approaches that consider both audio and words/tags associated with audio do not employ text processing models that are capable to generalize to tags unknown during training. In this work we propose a method for learning audio representations using an audio autoencoder (AAE), a general word embeddings model (WEM), and a multi-head self-attention (MHA) mechanism. MHA attends on the output of the WEM, providing a contextualized representation of the tags associated with the audio, and we align the output of MHA with the output of the encoder of AAE using a contrastive loss. We jointly optimize AAE and MHA and we evaluate the audio representations (i.e. the output of the encoder of AAE) by utilizing them in three different downstream tasks, namely sound, music genre, and music instrument classification. Our results show that employing multi-head self-attention with multiple heads in the tag-based network can induce better learned audio representations.