Music Playlist Title Generation: A Machine-Translation Approach
This work addresses a domain-specific problem for music streaming services by generating playlist titles, but it is incremental as it adapts existing machine-translation methods to a new application.
The authors tackled the problem of automatically generating playlist titles from music tracks using a machine-translation approach, adapting sequence-to-sequence models like RNN and Transformer to handle orderless track sequences. Results showed that Transformer models outperformed RNN, and techniques to remove input order, such as data augmentation and deleting positional encoding, further improved performance.
We propose a machine-translation approach to automatically generate a playlist title from a set of music tracks. We take a sequence of track IDs as input and a sequence of words in a playlist title as output, adapting the sequence-to-sequence framework based on Recurrent Neural Network (RNN) and Transformer to the music data. Considering the orderless nature of music tracks in a playlist, we propose two techniques that remove the order of the input sequence. One is data augmentation by shuffling and the other is deleting the positional encoding. We also reorganize the existing music playlist datasets to generate phrase-level playlist titles. The result shows that the Transformer models generally outperform the RNN model. Also, removing the order of input sequence improves the performance further.