WeSinger: Data-augmented Singing Voice Synthesis with Auxiliary Losses
In this paper, we develop a new multi-singer Chinese neural singing voice synthesis (SVS) system named WeSinger. To improve the accuracy and naturalness of synthesized singing voice, we design several specifical modules and techniques: 1) A deep bi-directional LSTM-based duration model with multi-sc...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2022-06 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Zhang, Zewang Zheng, Yibin Li, Xinhui Lu, Li |
description | In this paper, we develop a new multi-singer Chinese neural singing voice synthesis (SVS) system named WeSinger. To improve the accuracy and naturalness of synthesized singing voice, we design several specifical modules and techniques: 1) A deep bi-directional LSTM-based duration model with multi-scale rhythm loss and post-processing step; 2) A Transformer-alike acoustic model with progressive pitch-weighted decoder loss; 3) a 24 kHz pitch-aware LPCNet neural vocoder to produce high-quality singing waveforms; 4) A novel data augmentation method with multi-singer pre-training for stronger robustness and naturalness. To our knowledge, WeSinger is the first SVS system to adopt 24 kHz LPCNet and multi-singer pre-training simultaneously. Both quantitative and qualitative evaluation results demonstrate the effectiveness of WeSinger in terms of accuracy and naturalness, and WeSinger achieves state-of-the-art performance on the recent public Chinese singing corpus Opencpop\footnote{https://wenet.org.cn/opencpop/}. Some synthesized singing samples are available online\footnote{https://zzw922cn.github.io/wesinger/}. |
format | Article |
fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2641676384</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2641676384</sourcerecordid><originalsourceid>FETCH-proquest_journals_26416763843</originalsourceid><addsrcrecordid>eNqNitEKgjAYRkcQJOU7DLoWdNMp3UUlXXRn1KWM-tOJbbV_o3z7DHqA4IMD53wTEjDOk6hIGZuRELGL45iJnGUZD0h5hkrpBuyKbqWTkfTNHbSDK_3qcfRk1AVoNWjXAiqkL-VauvZv1StpB3owiIALMr3JHiH8cU6W5e642UcPa54e0NWd8VaPqWYiTUQueJHy_14fqkg7RA</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2641676384</pqid></control><display><type>article</type><title>WeSinger: Data-augmented Singing Voice Synthesis with Auxiliary Losses</title><source>Open Access: Freely Accessible Journals by multiple vendors</source><creator>Zhang, Zewang ; Zheng, Yibin ; Li, Xinhui ; Lu, Li</creator><creatorcontrib>Zhang, Zewang ; Zheng, Yibin ; Li, Xinhui ; Lu, Li</creatorcontrib><description>In this paper, we develop a new multi-singer Chinese neural singing voice synthesis (SVS) system named WeSinger. To improve the accuracy and naturalness of synthesized singing voice, we design several specifical modules and techniques: 1) A deep bi-directional LSTM-based duration model with multi-scale rhythm loss and post-processing step; 2) A Transformer-alike acoustic model with progressive pitch-weighted decoder loss; 3) a 24 kHz pitch-aware LPCNet neural vocoder to produce high-quality singing waveforms; 4) A novel data augmentation method with multi-singer pre-training for stronger robustness and naturalness. To our knowledge, WeSinger is the first SVS system to adopt 24 kHz LPCNet and multi-singer pre-training simultaneously. Both quantitative and qualitative evaluation results demonstrate the effectiveness of WeSinger in terms of accuracy and naturalness, and WeSinger achieves state-of-the-art performance on the recent public Chinese singing corpus Opencpop\footnote{https://wenet.org.cn/opencpop/}. Some synthesized singing samples are available online\footnote{https://zzw922cn.github.io/wesinger/}.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Singers ; Singing ; Synthesis ; Voice ; Waveforms</subject><ispartof>arXiv.org, 2022-06</ispartof><rights>2022. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</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>776,780</link.rule.ids></links><search><creatorcontrib>Zhang, Zewang</creatorcontrib><creatorcontrib>Zheng, Yibin</creatorcontrib><creatorcontrib>Li, Xinhui</creatorcontrib><creatorcontrib>Lu, Li</creatorcontrib><title>WeSinger: Data-augmented Singing Voice Synthesis with Auxiliary Losses</title><title>arXiv.org</title><description>In this paper, we develop a new multi-singer Chinese neural singing voice synthesis (SVS) system named WeSinger. To improve the accuracy and naturalness of synthesized singing voice, we design several specifical modules and techniques: 1) A deep bi-directional LSTM-based duration model with multi-scale rhythm loss and post-processing step; 2) A Transformer-alike acoustic model with progressive pitch-weighted decoder loss; 3) a 24 kHz pitch-aware LPCNet neural vocoder to produce high-quality singing waveforms; 4) A novel data augmentation method with multi-singer pre-training for stronger robustness and naturalness. To our knowledge, WeSinger is the first SVS system to adopt 24 kHz LPCNet and multi-singer pre-training simultaneously. Both quantitative and qualitative evaluation results demonstrate the effectiveness of WeSinger in terms of accuracy and naturalness, and WeSinger achieves state-of-the-art performance on the recent public Chinese singing corpus Opencpop\footnote{https://wenet.org.cn/opencpop/}. Some synthesized singing samples are available online\footnote{https://zzw922cn.github.io/wesinger/}.</description><subject>Singers</subject><subject>Singing</subject><subject>Synthesis</subject><subject>Voice</subject><subject>Waveforms</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNitEKgjAYRkcQJOU7DLoWdNMp3UUlXXRn1KWM-tOJbbV_o3z7DHqA4IMD53wTEjDOk6hIGZuRELGL45iJnGUZD0h5hkrpBuyKbqWTkfTNHbSDK_3qcfRk1AVoNWjXAiqkL-VauvZv1StpB3owiIALMr3JHiH8cU6W5e642UcPa54e0NWd8VaPqWYiTUQueJHy_14fqkg7RA</recordid><startdate>20220625</startdate><enddate>20220625</enddate><creator>Zhang, Zewang</creator><creator>Zheng, Yibin</creator><creator>Li, Xinhui</creator><creator>Lu, Li</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20220625</creationdate><title>WeSinger: Data-augmented Singing Voice Synthesis with Auxiliary Losses</title><author>Zhang, Zewang ; Zheng, Yibin ; Li, Xinhui ; Lu, Li</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_26416763843</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Singers</topic><topic>Singing</topic><topic>Synthesis</topic><topic>Voice</topic><topic>Waveforms</topic><toplevel>online_resources</toplevel><creatorcontrib>Zhang, Zewang</creatorcontrib><creatorcontrib>Zheng, Yibin</creatorcontrib><creatorcontrib>Li, Xinhui</creatorcontrib><creatorcontrib>Lu, Li</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>ProQuest Engineering Database</collection><collection>Publicly Available Content (ProQuest)</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Zhang, Zewang</au><au>Zheng, Yibin</au><au>Li, Xinhui</au><au>Lu, Li</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>WeSinger: Data-augmented Singing Voice Synthesis with Auxiliary Losses</atitle><jtitle>arXiv.org</jtitle><date>2022-06-25</date><risdate>2022</risdate><eissn>2331-8422</eissn><abstract>In this paper, we develop a new multi-singer Chinese neural singing voice synthesis (SVS) system named WeSinger. To improve the accuracy and naturalness of synthesized singing voice, we design several specifical modules and techniques: 1) A deep bi-directional LSTM-based duration model with multi-scale rhythm loss and post-processing step; 2) A Transformer-alike acoustic model with progressive pitch-weighted decoder loss; 3) a 24 kHz pitch-aware LPCNet neural vocoder to produce high-quality singing waveforms; 4) A novel data augmentation method with multi-singer pre-training for stronger robustness and naturalness. To our knowledge, WeSinger is the first SVS system to adopt 24 kHz LPCNet and multi-singer pre-training simultaneously. Both quantitative and qualitative evaluation results demonstrate the effectiveness of WeSinger in terms of accuracy and naturalness, and WeSinger achieves state-of-the-art performance on the recent public Chinese singing corpus Opencpop\footnote{https://wenet.org.cn/opencpop/}. Some synthesized singing samples are available online\footnote{https://zzw922cn.github.io/wesinger/}.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | EISSN: 2331-8422 |
ispartof | arXiv.org, 2022-06 |
issn | 2331-8422 |
language | eng |
recordid | cdi_proquest_journals_2641676384 |
source | Open Access: Freely Accessible Journals by multiple vendors |
subjects | Singers Singing Synthesis Voice Waveforms |
title | WeSinger: Data-augmented Singing Voice Synthesis with Auxiliary Losses |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-07T17%3A24%3A28IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=WeSinger:%20Data-augmented%20Singing%20Voice%20Synthesis%20with%20Auxiliary%20Losses&rft.jtitle=arXiv.org&rft.au=Zhang,%20Zewang&rft.date=2022-06-25&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2641676384%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2641676384&rft_id=info:pmid/&rfr_iscdi=true |