MMS-MSG: A Multi-purpose Multi-Speaker Mixture Signal Generator

The scope of speech enhancement has changed from a monolithic view of single, independent tasks, to a joint processing of complex conversational speech recordings. Training and evaluation of these single tasks requires synthetic data with access to intermediate signals that is as close as possible t...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cord-Landwehr, Tobias, von Neumann, Thilo, Boeddeker, Christoph, Haeb-Umbach, Reinhold
Format: Artikel
Sprache:eng
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Cord-Landwehr, Tobias
von Neumann, Thilo
Boeddeker, Christoph
Haeb-Umbach, Reinhold
description The scope of speech enhancement has changed from a monolithic view of single, independent tasks, to a joint processing of complex conversational speech recordings. Training and evaluation of these single tasks requires synthetic data with access to intermediate signals that is as close as possible to the evaluation scenario. As such data often is not available, many works instead use specialized databases for the training of each system component, e.g WSJ0-mix for source separation. We present a Multi-purpose Multi-Speaker Mixture Signal Generator (MMS-MSG) for generating a variety of speech mixture signals based on any speech corpus, ranging from classical anechoic mixtures (e.g., WSJ0-mix) over reverberant mixtures (e.g., SMS-WSJ) to meeting-style data. Its highly modular and flexible structure allows for the simulation of diverse environments and dynamic mixing, while simultaneously enabling an easy extension and modification to generate new scenarios and mixture types. These meetings can be used for prototyping, evaluation, or training purposes. We provide example evaluation data and baseline results for meetings based on the WSJ corpus. Further, we demonstrate the usefulness for realistic scenarios by using MMS-MSG to provide training data for the LibriCSS database.
doi_str_mv 10.48550/arxiv.2209.11494
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2209_11494</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2209_11494</sourcerecordid><originalsourceid>FETCH-LOGICAL-a674-94684d2325dcc0a54aeb6eb9c0d91c47c98dbb47ac793bbfc05b4fb81ed8a733</originalsourceid><addsrcrecordid>eNotz71OwzAYhWEvDKhwAUz4Bhzs-HNss6CqgoDUqIPZo88_QRahjdwElbtHlE5H73Kkh5A7wSswSvEHLKf8XdU1t5UQYOGaPHWdY51rH-madss4ZzYtZToc06XclPAzFdrl07yURF3-2ONI27RPBedDuSFXA47HdHvZFXEvz--bV7bdtW-b9ZZho4FZaAzEWtYqhsBRASbfJG8Dj1YE0MGa6D1oDNpK74fAlYfBG5GiQS3litz_v54B_VTyF5af_g_SnyHyF9gsQv4</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>MMS-MSG: A Multi-purpose Multi-Speaker Mixture Signal Generator</title><source>arXiv.org</source><creator>Cord-Landwehr, Tobias ; von Neumann, Thilo ; Boeddeker, Christoph ; Haeb-Umbach, Reinhold</creator><creatorcontrib>Cord-Landwehr, Tobias ; von Neumann, Thilo ; Boeddeker, Christoph ; Haeb-Umbach, Reinhold</creatorcontrib><description>The scope of speech enhancement has changed from a monolithic view of single, independent tasks, to a joint processing of complex conversational speech recordings. Training and evaluation of these single tasks requires synthetic data with access to intermediate signals that is as close as possible to the evaluation scenario. As such data often is not available, many works instead use specialized databases for the training of each system component, e.g WSJ0-mix for source separation. We present a Multi-purpose Multi-Speaker Mixture Signal Generator (MMS-MSG) for generating a variety of speech mixture signals based on any speech corpus, ranging from classical anechoic mixtures (e.g., WSJ0-mix) over reverberant mixtures (e.g., SMS-WSJ) to meeting-style data. Its highly modular and flexible structure allows for the simulation of diverse environments and dynamic mixing, while simultaneously enabling an easy extension and modification to generate new scenarios and mixture types. These meetings can be used for prototyping, evaluation, or training purposes. We provide example evaluation data and baseline results for meetings based on the WSJ corpus. Further, we demonstrate the usefulness for realistic scenarios by using MMS-MSG to provide training data for the LibriCSS database.</description><identifier>DOI: 10.48550/arxiv.2209.11494</identifier><language>eng</language><creationdate>2022-09</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2209.11494$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2209.11494$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Cord-Landwehr, Tobias</creatorcontrib><creatorcontrib>von Neumann, Thilo</creatorcontrib><creatorcontrib>Boeddeker, Christoph</creatorcontrib><creatorcontrib>Haeb-Umbach, Reinhold</creatorcontrib><title>MMS-MSG: A Multi-purpose Multi-Speaker Mixture Signal Generator</title><description>The scope of speech enhancement has changed from a monolithic view of single, independent tasks, to a joint processing of complex conversational speech recordings. Training and evaluation of these single tasks requires synthetic data with access to intermediate signals that is as close as possible to the evaluation scenario. As such data often is not available, many works instead use specialized databases for the training of each system component, e.g WSJ0-mix for source separation. We present a Multi-purpose Multi-Speaker Mixture Signal Generator (MMS-MSG) for generating a variety of speech mixture signals based on any speech corpus, ranging from classical anechoic mixtures (e.g., WSJ0-mix) over reverberant mixtures (e.g., SMS-WSJ) to meeting-style data. Its highly modular and flexible structure allows for the simulation of diverse environments and dynamic mixing, while simultaneously enabling an easy extension and modification to generate new scenarios and mixture types. These meetings can be used for prototyping, evaluation, or training purposes. We provide example evaluation data and baseline results for meetings based on the WSJ corpus. Further, we demonstrate the usefulness for realistic scenarios by using MMS-MSG to provide training data for the LibriCSS database.</description><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz71OwzAYhWEvDKhwAUz4Bhzs-HNss6CqgoDUqIPZo88_QRahjdwElbtHlE5H73Kkh5A7wSswSvEHLKf8XdU1t5UQYOGaPHWdY51rH-madss4ZzYtZToc06XclPAzFdrl07yURF3-2ONI27RPBedDuSFXA47HdHvZFXEvz--bV7bdtW-b9ZZho4FZaAzEWtYqhsBRASbfJG8Dj1YE0MGa6D1oDNpK74fAlYfBG5GiQS3litz_v54B_VTyF5af_g_SnyHyF9gsQv4</recordid><startdate>20220923</startdate><enddate>20220923</enddate><creator>Cord-Landwehr, Tobias</creator><creator>von Neumann, Thilo</creator><creator>Boeddeker, Christoph</creator><creator>Haeb-Umbach, Reinhold</creator><scope>GOX</scope></search><sort><creationdate>20220923</creationdate><title>MMS-MSG: A Multi-purpose Multi-Speaker Mixture Signal Generator</title><author>Cord-Landwehr, Tobias ; von Neumann, Thilo ; Boeddeker, Christoph ; Haeb-Umbach, Reinhold</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a674-94684d2325dcc0a54aeb6eb9c0d91c47c98dbb47ac793bbfc05b4fb81ed8a733</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><toplevel>online_resources</toplevel><creatorcontrib>Cord-Landwehr, Tobias</creatorcontrib><creatorcontrib>von Neumann, Thilo</creatorcontrib><creatorcontrib>Boeddeker, Christoph</creatorcontrib><creatorcontrib>Haeb-Umbach, Reinhold</creatorcontrib><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Cord-Landwehr, Tobias</au><au>von Neumann, Thilo</au><au>Boeddeker, Christoph</au><au>Haeb-Umbach, Reinhold</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>MMS-MSG: A Multi-purpose Multi-Speaker Mixture Signal Generator</atitle><date>2022-09-23</date><risdate>2022</risdate><abstract>The scope of speech enhancement has changed from a monolithic view of single, independent tasks, to a joint processing of complex conversational speech recordings. Training and evaluation of these single tasks requires synthetic data with access to intermediate signals that is as close as possible to the evaluation scenario. As such data often is not available, many works instead use specialized databases for the training of each system component, e.g WSJ0-mix for source separation. We present a Multi-purpose Multi-Speaker Mixture Signal Generator (MMS-MSG) for generating a variety of speech mixture signals based on any speech corpus, ranging from classical anechoic mixtures (e.g., WSJ0-mix) over reverberant mixtures (e.g., SMS-WSJ) to meeting-style data. Its highly modular and flexible structure allows for the simulation of diverse environments and dynamic mixing, while simultaneously enabling an easy extension and modification to generate new scenarios and mixture types. These meetings can be used for prototyping, evaluation, or training purposes. We provide example evaluation data and baseline results for meetings based on the WSJ corpus. Further, we demonstrate the usefulness for realistic scenarios by using MMS-MSG to provide training data for the LibriCSS database.</abstract><doi>10.48550/arxiv.2209.11494</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2209.11494
ispartof
issn
language eng
recordid cdi_arxiv_primary_2209_11494
source arXiv.org
title MMS-MSG: A Multi-purpose Multi-Speaker Mixture Signal Generator
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-14T19%3A23%3A47IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=MMS-MSG:%20A%20Multi-purpose%20Multi-Speaker%20Mixture%20Signal%20Generator&rft.au=Cord-Landwehr,%20Tobias&rft.date=2022-09-23&rft_id=info:doi/10.48550/arxiv.2209.11494&rft_dat=%3Carxiv_GOX%3E2209_11494%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true