通过视觉场景驱动扩散进行声学匹配和消除混响的相互学习

Jian Ma, Wenguan Wang, Yi Yang, Feng Zheng
{"title":"通过视觉场景驱动扩散进行声学匹配和消除混响的相互学习","authors":"Jian Ma, Wenguan Wang, Yi Yang, Feng Zheng","doi":"arxiv-2407.10373","DOIUrl":null,"url":null,"abstract":"Visual acoustic matching (VAM) is pivotal for enhancing the immersive\nexperience, and the task of dereverberation is effective in improving audio\nintelligibility. Existing methods treat each task independently, overlooking\nthe inherent reciprocity between them. Moreover, these methods depend on paired\ntraining data, which is challenging to acquire, impeding the utilization of\nextensive unpaired data. In this paper, we introduce MVSD, a mutual learning\nframework based on diffusion models. MVSD considers the two tasks\nsymmetrically, exploiting the reciprocal relationship to facilitate learning\nfrom inverse tasks and overcome data scarcity. Furthermore, we employ the\ndiffusion model as foundational conditional converters to circumvent the\ntraining instability and over-smoothing drawbacks of conventional GAN\narchitectures. Specifically, MVSD employs two converters: one for VAM called\nreverberator and one for dereverberation called dereverberator. The\ndereverberator judges whether the reverberation audio generated by reverberator\nsounds like being in the conditional visual scenario, and vice versa. By\nforming a closed loop, these two converters can generate informative feedback\nsignals to optimize the inverse tasks, even with easily acquired one-way\nunpaired data. Extensive experiments on two standard benchmarks, i.e.,\nSoundSpaces-Speech and Acoustic AVSpeech, exhibit that our framework can\nimprove the performance of the reverberator and dereverberator and better match\nspecified visual scenarios.","PeriodicalId":501178,"journal":{"name":"arXiv - CS - Sound","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Mutual Learning for Acoustic Matching and Dereverberation via Visual Scene-driven Diffusion\",\"authors\":\"Jian Ma, Wenguan Wang, Yi Yang, Feng Zheng\",\"doi\":\"arxiv-2407.10373\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Visual acoustic matching (VAM) is pivotal for enhancing the immersive\\nexperience, and the task of dereverberation is effective in improving audio\\nintelligibility. Existing methods treat each task independently, overlooking\\nthe inherent reciprocity between them. Moreover, these methods depend on paired\\ntraining data, which is challenging to acquire, impeding the utilization of\\nextensive unpaired data. In this paper, we introduce MVSD, a mutual learning\\nframework based on diffusion models. MVSD considers the two tasks\\nsymmetrically, exploiting the reciprocal relationship to facilitate learning\\nfrom inverse tasks and overcome data scarcity. Furthermore, we employ the\\ndiffusion model as foundational conditional converters to circumvent the\\ntraining instability and over-smoothing drawbacks of conventional GAN\\narchitectures. Specifically, MVSD employs two converters: one for VAM called\\nreverberator and one for dereverberation called dereverberator. The\\ndereverberator judges whether the reverberation audio generated by reverberator\\nsounds like being in the conditional visual scenario, and vice versa. By\\nforming a closed loop, these two converters can generate informative feedback\\nsignals to optimize the inverse tasks, even with easily acquired one-way\\nunpaired data. Extensive experiments on two standard benchmarks, i.e.,\\nSoundSpaces-Speech and Acoustic AVSpeech, exhibit that our framework can\\nimprove the performance of the reverberator and dereverberator and better match\\nspecified visual scenarios.\",\"PeriodicalId\":501178,\"journal\":{\"name\":\"arXiv - CS - Sound\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-07-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Sound\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2407.10373\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Sound","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2407.10373","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

视觉声学匹配(VAM)对于增强身临其境的体验至关重要,而消除混响的任务则能有效提高音频的可理解性。现有方法将这两项任务分开处理,忽略了它们之间固有的互惠性。此外,这些方法依赖于配对训练数据,而配对训练数据的获取极具挑战性,阻碍了对大量非配对数据的利用。本文介绍了基于扩散模型的互学框架 MVSD。MVSD 对称考虑两个任务,利用互惠关系促进逆任务学习,克服数据稀缺问题。此外,我们采用扩散模型作为基础条件转换器,以规避传统 GAN 架构的训练不稳定性和过度平滑缺点。具体来说,MVSD 采用了两个转换器:一个用于 VAM,称为转换器(reverberator);另一个用于消除混响,称为消除混响器(dereverberator)。去混响器判断混响器生成的混响音频是否听起来像在条件视觉场景中,反之亦然。通过形成闭环,这两个转换器可以生成信息反馈信号,以优化逆任务,即使是轻松获取的单向非配对数据也不例外。在两个标准基准(即 SoundSpaces-Speech 和 Acoustic AVSpeech)上进行的广泛实验表明,我们的框架可以提高混响器和消除混响器的性能,并更好地匹配指定的视觉场景。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Mutual Learning for Acoustic Matching and Dereverberation via Visual Scene-driven Diffusion
Visual acoustic matching (VAM) is pivotal for enhancing the immersive experience, and the task of dereverberation is effective in improving audio intelligibility. Existing methods treat each task independently, overlooking the inherent reciprocity between them. Moreover, these methods depend on paired training data, which is challenging to acquire, impeding the utilization of extensive unpaired data. In this paper, we introduce MVSD, a mutual learning framework based on diffusion models. MVSD considers the two tasks symmetrically, exploiting the reciprocal relationship to facilitate learning from inverse tasks and overcome data scarcity. Furthermore, we employ the diffusion model as foundational conditional converters to circumvent the training instability and over-smoothing drawbacks of conventional GAN architectures. Specifically, MVSD employs two converters: one for VAM called reverberator and one for dereverberation called dereverberator. The dereverberator judges whether the reverberation audio generated by reverberator sounds like being in the conditional visual scenario, and vice versa. By forming a closed loop, these two converters can generate informative feedback signals to optimize the inverse tasks, even with easily acquired one-way unpaired data. Extensive experiments on two standard benchmarks, i.e., SoundSpaces-Speech and Acoustic AVSpeech, exhibit that our framework can improve the performance of the reverberator and dereverberator and better match specified visual scenarios.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Benchmarking Sub-Genre Classification For Mainstage Dance Music PDAF: A Phonetic Debiasing Attention Framework For Speaker Verification Evaluation of real-time transcriptions using end-to-end ASR models Machine Anomalous Sound Detection Using Spectral-temporal Modulation Representations Derived from Machine-specific Filterbanks Harmonic Reasoning in Large Language Models
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1