Contrastive Learning-based Audio to Lyrics Alignment for Multiple Languages

Abstract

Lyrics alignment gained considerable attention in recent years. State-of-the-art systems either re-use established speech recognition toolkits, or design end-to-end solutions involving a Connectionist Temporal Classification (CTC) loss. However, both approaches suffer from specific weaknesses: toolkits are known for their complexity, and CTC systems use a loss designed for transcription which can limit alignment accuracy. In this paper, we use instead a contrastive learning procedure that derives cross-modal embeddings linking the audio and text domains. This way, we obtain a novel system that is simple to train end-to-end, can make use of weakly annotated training data, jointly learns a powerful text model, and is tailored to alignment. The system is not only the first to yield an average absolute error below 0.2 seconds on the standard Jamendo dataset but it is also robust to other languages, even when trained on English data only. Finally, we release word-level alignments for the JamendoLyrics Multi-Lang dataset.

Related

September 2023 | RecSys

Accelerating Creator Audience Building through Centralized Exploration

Buket Baran, Guilherme Dinis Junior, Antonina Danylenko, Olayinka S. Folorunso, Gösta Forsum, Maksym Lefarov, Lucas Maystre, Yu Zhao

August 2023 | Interspeech

Lightweight and Efficient Spoken Language Identification of Long-form Audio

Winstead Zhu, Md Iftekhar Tanveer, Yang Janet Liu, Seye Ojumu, Rosie Jones

July 2023 | KDD

Impatient Bandits: Optimizing for the Long-Term Without Delay

Thomas McDonald, Lucas Maystre, Mounia Lalmas, Daniel Russo, Kamil Ciosek