Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2510.00006

Unpacking Musical Symbolism in Online Communities: Content-Based and Network-Centric Approaches

Kajwan Ziaoddini
View original source ↗Read source document ↗Export metadata ↓

Abstract

This paper examines how musical symbolism is produced and circulated in online communities by combining content-based music analysis with a lightweight network perspective on lyrics. Using a curated corpus of 275 chart-topping songs enriched with audio descriptors (energy, danceability, loudness, liveness, valence, acousticness, speechiness, popularity) and full lyric transcripts, we build a reproducible pipeline that (i) quantifies temporal trends in sonic attributes, (ii) models lexical salience and co-occurrence, and (iii) profiles mood by genre. We find a decade-long decline in energy (79 -> 58) alongside a rise in danceability (59 -> 73); valence peaks in 2013 (63) and dips in 2014-2016 (42) before partially recovering. Correlation analysis shows strong coupling of energy with loudness (r = 0.74) and negative associations for acousticness with both energy (r = -0.54) and loudness (r = -0.51); danceability is largely orthogonal to other features (|r| < 0.20). Lyric tokenization (>114k tokens) reveals a pronoun-centric lexicon "I/you/me/my" and a dense co-occurrence structure in which interpersonal address anchors mainstream narratives. Mood differs systematically by style: R&B exhibits the highest mean valence (96), followed by K-Pop/Pop (77) and Indie/Pop (70), whereas Latin/Reggaeton is lower (37) despite high danceability. Read through a subcultural identity lens, these patterns suggest the mainstreaming of previously peripheral codes and a commercial preference for relaxed yet rhythmically engaging productions that sustain collective participation without maximal intensity. Methodologically, we contribute an integrated MIR-plus-network workflow spanning summary statistics, correlation structure, lexical co-occurrence matrices, and genre-wise mood profiling that is robust to modality sparsity and suitable for socially aware recommendation or community-level diffusion studies.

Explore related subjects

cs.SDcs.CLcs.CYcs.MMeess.AS

Keep this discovery

BibTeXRIS

Kajwan Ziaoddini. 2025-09-13. Unpacking Musical Symbolism in Online Communities: Content-Based and Network-Centric Approaches. https://arxiv.org/abs/2510.00006

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Xiaomi-CocktailASR-1 Technical Report

Recently, large language model (LLM) based ASR models have achieved significant progress, yet they generally lack support for multi-speaker scenarios, where the cocktail party problem remains a critical bottleneck for further advancing ASR. Existing TS-ASR methods, including end-to-end architectures with speaker embeddings and latest LLM-based explorations suffer from degraded single-speaker performance and the inability to reject when the target speaker is absent. In this paper, we propose Xiaomi-CocktailASR-1, an LLM-based end-to-end TS-ASR architecture. By utilizing reference speech as voiceprint prompts, it directly transcribes the target speaker's speech without requiring speech separation. Xiaomi-CocktailASR-1 maintains competitive performance in single-speaker scenarios, comparable to mainstream ASR models. It also features a negative sample rejection capability, outputting empty text when the target speaker is absent from the mixed speech. Additionally, Xiaomi-CocktailASR-1 supports a Chain-of-Thought (CoT) reasoning mode to provide explicit reasoning steps. Extensive experiments on various synthetic and real-world multispeaker benchmarks demonstrate that Xiaomi-CocktailASR-1 achieves state-of-the-art performance, effectively addressing the cocktail party problem through a unified architecture that balances multispeaker and single-speaker recognition accuracy, along with rejection capability.

cs.SD↗
arXiv2026

EConv-TasNet: Efficient Conv-TasNet for Effective Speech Separation

Conv-TasNet has served as a strong baseline for time-domain speech separation, and many studies have extended it with advanced architectures such as dual-path networks, U-Nets, and attention mechanisms. However, these methods often introduce high computational cost and complexity, limiting their deployment in resource-constrained scenarios. To address this issue, we propose eConv-TasNet, an efficient variant of Conv-TasNet that improves both effectiveness and efficiency without relying on resource-intensive modules. The proposed model consists of a group-wise early-splitting (GES) module and a multi-group feature aggregation (MGFA) module. GES generates discriminative speaker embeddings at intermediate stages, while MGFA progressively aggregates these group-level representations for refined mask estimation. Experimental results show that eConv-TasNet reduces model size by 22.4%, accelerates inference by 18.9%, and improves SI-SNRi by 14.0%-28.0% across three public benchmarks. Moreover, it achieves competitive performance compared with state-of-the-art methods while requiring significantly fewer parameters and lower inference cost. These results demonstrate a favorable efficiency-effectiveness trade-off for edge deployment.

cs.SD↗