Benchmarking the First Realistic Dataset for Speech Separation

This paper presents a thorough benchmarking analysis of a recently introduced realistic dataset for speech separation tasks. The dataset contains audio mixtures that replicate real-life scenarios and is accompanied by ground truths, making it a valuable resource for researchers. Although the datase...

Full description

Saved in:
Bibliographic Details
Main Authors: Rawad MELHEM, Oumayma AL DAKKAK, Assef JAFAR
Format: Article
Language:English
Published: Institute of Fundamental Technological Research Polish Academy of Sciences 2025-07-01
Series:Archives of Acoustics
Subjects:
Online Access:https://acoustics.ippt.pan.pl/index.php/aa/article/view/4180
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849695817537945600
author Rawad MELHEM
Oumayma AL DAKKAK
Assef JAFAR
author_facet Rawad MELHEM
Oumayma AL DAKKAK
Assef JAFAR
author_sort Rawad MELHEM
collection DOAJ
description This paper presents a thorough benchmarking analysis of a recently introduced realistic dataset for speech separation tasks. The dataset contains audio mixtures that replicate real-life scenarios and is accompanied by ground truths, making it a valuable resource for researchers. Although the dataset construction methodology was recently disclosed, its benchmarking and detailed performance analysis have not yet been conducted. In this study, we evaluate the performance of four speech separation models using two distinct testing sets, ensuring a robust evaluation. Our findings underscore the dataset’s efficacy to advance speech separation research within authentic environments. Furthermore, we propose a novel approach for assessing metrics in real-world speech separation systems, where ground truths are unavailable. This method aims to improve accuracy evaluations and refine models for practical applications.We make the dataset publicly available to encourage innovation and collaboration in the field.
format Article
id doaj-art-76e1c85f750c4f87b4ea6533f641d21f
institution DOAJ
issn 0137-5075
2300-262X
language English
publishDate 2025-07-01
publisher Institute of Fundamental Technological Research Polish Academy of Sciences
record_format Article
series Archives of Acoustics
spelling doaj-art-76e1c85f750c4f87b4ea6533f641d21f2025-08-20T03:19:39ZengInstitute of Fundamental Technological Research Polish Academy of SciencesArchives of Acoustics0137-50752300-262X2025-07-0110.24425/aoa.2025.154816Benchmarking the First Realistic Dataset for Speech SeparationRawad MELHEM0Oumayma AL DAKKAK1Assef JAFAR2Higher Institute for Applied Sciences and TechnologyHigher Institute for Applied Sciences and TechnologyHigher Institute for Applied Sciences and Technology This paper presents a thorough benchmarking analysis of a recently introduced realistic dataset for speech separation tasks. The dataset contains audio mixtures that replicate real-life scenarios and is accompanied by ground truths, making it a valuable resource for researchers. Although the dataset construction methodology was recently disclosed, its benchmarking and detailed performance analysis have not yet been conducted. In this study, we evaluate the performance of four speech separation models using two distinct testing sets, ensuring a robust evaluation. Our findings underscore the dataset’s efficacy to advance speech separation research within authentic environments. Furthermore, we propose a novel approach for assessing metrics in real-world speech separation systems, where ground truths are unavailable. This method aims to improve accuracy evaluations and refine models for practical applications.We make the dataset publicly available to encourage innovation and collaboration in the field. https://acoustics.ippt.pan.pl/index.php/aa/article/view/4180single-channelspeech separationdeep learningcorpusdatasets
spellingShingle Rawad MELHEM
Oumayma AL DAKKAK
Assef JAFAR
Benchmarking the First Realistic Dataset for Speech Separation
Archives of Acoustics
single-channel
speech separation
deep learning
corpus
datasets
title Benchmarking the First Realistic Dataset for Speech Separation
title_full Benchmarking the First Realistic Dataset for Speech Separation
title_fullStr Benchmarking the First Realistic Dataset for Speech Separation
title_full_unstemmed Benchmarking the First Realistic Dataset for Speech Separation
title_short Benchmarking the First Realistic Dataset for Speech Separation
title_sort benchmarking the first realistic dataset for speech separation
topic single-channel
speech separation
deep learning
corpus
datasets
url https://acoustics.ippt.pan.pl/index.php/aa/article/view/4180
work_keys_str_mv AT rawadmelhem benchmarkingthefirstrealisticdatasetforspeechseparation
AT oumaymaaldakkak benchmarkingthefirstrealisticdatasetforspeechseparation
AT assefjafar benchmarkingthefirstrealisticdatasetforspeechseparation