SearcharxivSearch

arXiv · 2202.05539

A Sonification of the zCOSMOS Galaxy Dataset

Abstract

Sonification is the transformation of data into acoustic signals, achievable through different techniques. Sonification can be defined as a way to represent data values and relations as perceivable sounds, aiming at facilitating their communication and interpretation. Like data visualization provides meaning via images, sonification conveys meaning via sound. Sonification approaches are useful in a number of scenario. A first case is the possibility to receive information while keeping other sensory channels free, like in medical environment, in driving experience, etc. Another scenario addresses an easier recognition of patterns when data present high dimensionality and cardinality. Finally, sonification can be applied to presentation and dissemination initiatives, also with artistic goals. The zCOSMOS dataset contains detailed data about almost 20000 galaxies, describing the evolution of a relatively small portion of the universe in the last 10 million years in terms of galaxy mass, absolute luminosity, redshift, distance, age, and star formation rate. The present paper proposes a sonification for the mentioned dataset, with the following goals: i) providing a general description of the dataset, accessible via sound, which could also make unnoticed patterns emerge; ii) realizing an artistic but scientifically accurate sonic portrait of a portion of the universe, thus filling the gap between art and science in the context of scientific dissemination and so-called "edutainment"; iii) adding value to the dataset, since also scientific data and achievements must be considered as a cultural heritage that needs to be preserved and enhanced. Both scientific and technological aspects of the sonification are addressed.

Explore related subjects

Keep this discovery

BibTeXRIS

S. Bardelli, Claudia Ferretti, Luca Andrea Ludovico, Giorgio Presti, Maurizio Rinaldi. 2022-02-11. A Sonification of the zCOSMOS Galaxy Dataset. https://doi.org/10.22201/ia.14052059p.2022.54.10

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

KEEP EXPLORING

Related papers

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

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