Contrastive Learning-based Audio to Lyrics Alignment for Multiple Languages
Simon Durand, Daniel Stoller, Sebastian Ewert
Audio intelligence research at Spotify is advancing the state of the art in understanding music at scale to enhance how it is created, identified and consumed. We build bridges from raw audio to description, similarity, recommendation and music creation by developing machine listening technologies and synthesis algorithms. These power the next-generation of differentiating products and experiences, blurring the line between creators and consumers. Examples of active research areas in audio intelligence include information retrieval, source separation, auto tagging, auto mixing, mashups, sound modeling, vocals characterization, and music promotion.
Simon Durand, Daniel Stoller, Sebastian Ewert
M Iftekhar Tanveer, Diego Casabuena, Jussi Karlgren, Rosie Jones
Katariina Martikainen, Jussi Karlgren, Khiet Truong