Song Lyrics Summarization Inspired by Audio Thumbnailing

Fell, Michael; Cabrio, Elena; Gandon, Fabien; Giboin, Alain

doi:10.26615/978-954-452-056-4_038

Cited by 6 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is toggled by clicking on the violet-blue square on top of the song text. For a subset of songs the color opacity indicates how repetitive and representative a segment is, based on the fitness metric that we proposed in [25]. Note how in Figure 4, the segments 2, 4 and 7 are shaded more darkly than the surrounding ones.…”

Section: Lyrics Structure Annotationsmentioning

confidence: 99%

“…Given the repeating forms, peculiar structure and other unique characteristics of song lyrics, in [25] we introduced a method for extractive summarization of lyrics that takes advantage of these additional elements to more accurately identify relevant information in song lyrics. More specifically, it relies on the intimate relationship between the audio and the lyrics.…”

Section: Lyrics Summarymentioning

confidence: 99%

See 1 more Smart Citation

The WASABI song corpus and knowledge graph for music lyrics analysis

Fell

Cabrio²,

Tikat³

et al. 2022

Lang Resources & Evaluation

View full text Add to dashboard Cite

We present the WASABI 1 Song Corpus, a large corpus of songs enriched with metadata extracted from music databases on the Web, and resulting from the processing of song lyrics and from audio analysis. More specifically, given that lyrics encode an important part of the semantics of a song, we focus here on the description of the methods we proposed to extract relevant information from the lyrics, such as their structure segmentation, their topics, the explicitness of the lyrics content, the salient passages of a song and the emotions conveyed. The corpus contains 1.73M songs with lyrics (1.41M unique lyrics) annotated at different levels with the output of the above mentioned methods. The corpus labels and the provided methods can be exploited by music search engines and music professionals (e.g. journalists, radio presenters) to better handle large collections of lyrics, allowing an intelligent browsing, categorization and recommendation of songs. We demonstrate the utility and versatility of the WASABI Song Corpus in three concrete application scenarios. Together with the work on the corpus, we present the work achieved to transition the dataset into a knowledge graph, the WASABI RDF Knowledge Graph, and we show how this will enable an even richer set of applications.

show abstract

Section: Lyrics Structure Annotationsmentioning

confidence: 99%

Section: Lyrics Summarymentioning

confidence: 99%

The WASABI song corpus and knowledge graph for music lyrics analysis

Fell

Cabrio²,

Tikat³

et al. 2022

Lang Resources & Evaluation

View full text Add to dashboard Cite

show abstract

“…salient passages of a song: we introduced a method for extractive summarization of lyrics that relies on the intimate relationship between the audio and the lyrics (audio thumbnailing approach) [12]; emotions conveyed : we trained an emotion regression model using BERT to classify emotions in lyrics based on the valence-arousal model [13].…”

Section: Generating Lyrics Metadatamentioning

confidence: 99%

“…The ontology can be dereferenced with content negotiation, as well as all the terms of the ontology. It can be downloaded from the repository 12 where graphical visualizations are also available.…”

Section: The Wasabi Ontologymentioning

confidence: 99%

The WASABI Dataset: Cultural, Lyrics and Audio Analysis Metadata About 2 Million Popular Commercially Released Songs

Buffa

Cabrio²,

Fell³

et al. 2021

The Semantic Web

Self Cite

View full text Add to dashboard Cite

Since 2017, the goal of the two-million song WASABI database has been to build a knowledge graph linking collected metadata (artists, discography, producers, dates, etc.) with metadata generated by the analysis of both the songs' lyrics (topics, places, emotions, structure, etc.) and audio signal (chords, sound, etc.). It relies on natural language processing and machine learning methods for extraction, and semantic Web frameworks for representation and integration. It describes more than 2 millions commercial songs, 200K albums and 77K artists. It can be exploited by music search engines, music professionals (e.g. journalists, radio presenters, music teachers) or scientists willing to analyze popular music published since 1950. It is available under an open license, in multiple formats and with online and open source services including an interactive navigator, a REST API and a SPARQL endpoint.

show abstract

“…Lyrics summarization aims to preserve key information and the overall meaning of lyrics. As a special kind of text summarization task, Fell et al [9] propose to employ the generic text summarization models over lyrics.…”

Section: Related Workmentioning

confidence: 99%

Multi-modal Chorus Recognition for Improving Song Search

Wang

et al. 2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

We discuss a novel task, Chorus Recognition, which could potentially benefit downstream tasks such as song search and music summarization. Different from the existing tasks such as music summarization or lyrics summarization relying on single-modal information, this paper models chorus recognition as a multi-modal one by utilizing both the lyrics and the tune information of songs. We propose a multi-modal Chorus Recognition model that considers diverse features. Besides, we also create and publish the first Chorus Recognition dataset containing 627 songs for public use. Our empirical study performed on the dataset demonstrates that our approach outperforms several baselines in chorus recognition. In addition, our approach also helps to improve the accuracy of its downstream task -song search by more than 10.6%.

show abstract

Song Lyrics Summarization Inspired by Audio Thumbnailing

Cited by 6 publications

References 25 publications

The WASABI song corpus and knowledge graph for music lyrics analysis

The WASABI song corpus and knowledge graph for music lyrics analysis

The WASABI Dataset: Cultural, Lyrics and Audio Analysis Metadata About 2 Million Popular Commercially Released Songs

Multi-modal Chorus Recognition for Improving Song Search

Contact Info

Product

Resources

About