混ざった声を聞き分ける最新技術：音源分離と目的音声抽出
Digital data available（科学技術振興機構）
Begin reading now
J-STAGE
Holdings of Libraries in Japan
This page shows libraries in Japan other than the National Diet Library that hold the material.
Please contact your local library for information on how to use materials or whether it is possible to request materials from the holding libraries.
other
J-STAGE
DigitalCiNii Research
Search ServiceDigitalYou can check the holdings of institutions and databases with which CiNii Research is linked at the site of CiNii Research.
Bibliographic Record
You can check the details of this material, its authority (keywords that refer to materials on the same subject, author's name, etc.), etc.
- Material Type
- 記事
- Publication Date
- 2025-04-01
- Publication Date (W3CDTF)
- 2025-04-01
- Periodical title
- 電子情報通信学会 基礎・境界ソサイエティ ＦＵＮＤＡＭＥＮＴＡＬＳ ＲＥＶＩＥＷ
- No. or year of volume/issue
- 18 4
- Volume
- 18
- Issue
- 4
- Pages
- 267-278
- Publication date of volume/issue (W3CDTF)
- 2025-04-01
- Publication (Periodical Title)
- The Institute of Electronics, Information and Communication Engineers
- Text Language Code
- ja
- Subject Heading
- Target Audience
- 一般
- DOI
- 10.1587/essfr.18.4_267
- Related Material (URI)
- References
- Independent Vector Analysis via Log-Quadratically Penalized Quadratic MinimizationWavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech ProcessingIndependent Vector Extraction for Fast Joint Blind Source Separation and DereverberationSelf-Supervised Speech Representation Learning: A ReviewSpeech Enhancement and Dereverberation With Diffusion-Based Generative ModelsFast Online Source Steering Algorithm for Tracking Single Moving Source Using Online Independent Vector AnalysisBlind and Neural Network-Guided Convolutional Beamformer for Joint Denoising, Dereverberation, and Source SeparationImproving Speaker Discrimination of Target Speech Extraction With Time-Domain SpeakerbeamMicrophone Array Signal Processing and Deep Learning for Speech Enhancement: Combining model-based and data-driven approaches to parameter estimation and filteringTarget Speech Extraction with Conditional Diffusion ModelSpeaker Activity Driven Neural Speech ExtractionThe Conversation: Deep Audio-Visual Speech EnhancementSpeech Dereverberation Based on Maximum-Likelihood Estimation With Time-Varying Gaussian Source ModelJoint Dereverberation and Separation With Iterative Source SteeringISS2: An Extension of Iterative Source Steering Algorithm for Majorization-Minimization-Based Independent Vector AnalysisSpeakerBeam: Speaker Aware Neural Network for Target Speaker Extraction in Speech MixturesAutoregressive Moving Average Jointly-Diagonalizable Spatial Covariance Analysis for Joint Source Separation and DereverberationTarget-Speaker Voice Activity Detection: A Novel Approach for Multi-Speaker Diarization in a Dinner Party ScenarioTarget Speech Extraction with Pre-Trained Self-Supervised Learning ModelsMulti-Stream Diffusion Model for Probabilistic Integration of Model-Based and Data-Driven Speech EnhancementMajorization-Minimization Algorithms in Signal Processing, Communications, and Machine LearningAuxiliary-Function-Based Independent Component Analysis for Super-Gaussian SourcesSolution of Permutation Problem in Frequency Domain ICA, Using Multivariate Probability Density FunctionsMultichannel blind deconvolution and equalization using the natural gradientDetermined BSS Based on Time-Frequency Masking and Its Application to Harmonic Vector AnalysisGeneralization of Multi-Channel Linear Prediction Methods for Blind MIMO Impulse Response ShorteningSwitching Independent Vector Analysis and its Extension to Blind and Spatially Guided Convolutional Beamforming AlgorithmsBlind and Spatially-Regularized Online Joint Optimization of Source Separation, Dereverberation, and Noise ReductionTF-GridNet: Integrating Full- and Sub-Band Modeling for Speech SeparationNeural Blind Source Separation and Diarization for Distant Speech RecognitionNeural Target Speech Extraction: An overviewBUDDy: Single-Channel Blind Unsupervised Dereverberation with Diffusion ModelsAcoustic Modeling for Google HomeA Consolidated Perspective on Multimicrophone Speech Enhancement and Source SeparationFast and Stable Blind Source Separation with Rank-1 UpdatesSpeech DereverberationBeamforming: a versatile approach to spatial filteringAn auxiliary-function approach to online independent vector analysis for real-time blind source separationReal-Time Independent Vector Analysis for Convolutive Blind Source SeparationEnd-to-End SpeakerBeam for Single Channel Target Speech RecognitionEnd-to-End Multi-Speaker Speech Recognition Using Speaker Embeddings and Transfer LearningThe CHiME-8 DASR Challenge for Generalizable and Array Agnostic Distant Automatic Speech Recognition and DiarizationComputationally Efficient and Versatile Framework for Joint Optimization of Blind Speech Separation and DereverberationStreaming Target-Speaker ASR with Neural TransducerMasked Modeling Duo: Towards a Universal Audio Pre-Training FrameworkAudioLDM 2: Learning Holistic Audio Generation With Self-Supervised PretrainingICASSP 2023 Speech Signal Improvement ChallengePersonal VAD: Speaker-Conditioned Voice Activity DetectionMulti-Channel Linear Prediction-Based Speech Dereverberation With Sparse PriorsSpeaker-Aware Neural Network Based Beamformer for Speaker Extraction in Speech MixturesU-Net: Convolutional Networks for Biomedical Image SegmentationDeep clustering: Discriminative embeddings for segmentation and separationIndependent Low-Rank Matrix Analysis with Decorrelation LearningSEGAN: Speech Enhancement Generative Adversarial NetworkBlind separation of instantaneous mixtures of nonstationary sourcesLooking to listen at the cocktail partyFast fixed-point independent vector analysis algorithms for convolutive blind source separationOn Optimal Frequency-Domain Multichannel Linear Filtering for Noise ReductionSpeech Dereverberation Based on Variance-Normalized Delayed Linear PredictionIndependent component analysis, A new concept?Blind Source Separation Exploiting Higher-Order Frequency DependenciesA Robust and Precise Method for Solving the Permutation Problem of Frequency-Domain Blind Source SeparationDetermined Blind Source Separation Unifying Independent Vector Analysis and Nonnegative Matrix FactorizationLow-latency real-time blind source separation for hearing aids based on time-domain implementation of online independent vector analysis with truncation of non-causal componentsMultitalker Speech Separation With Utterance-Level Permutation Invariant Training of Deep Recurrent Neural NetworksJoint Separation and Dereverberation of Reverberant Mixtures with Determined Multichannel Non-Negative Matrix FactorizationBlind Separation and Dereverberation of Speech Mixtures by Joint OptimizationA summary of the REVERB challenge: state-of-the-art and remaining challenges in reverberant speech processing researchFast and robust fixed-point algorithms for independent component analysisInverse filtering of room acousticsConv-TasNet: Surpassing Ideal Time–Frequency Magnitude Masking for Speech Separation
- Data Provider (Database)
- 国立情報学研究所 : CiNii Research
- Original Data Provider (Database)
- Japan Link CenterCrossref
- Summary, etc.
- <p>In this paper, the latest advancements in source separation and target speech extraction technologies are reviewed. The former technology separates individual sounds from an acoustic signal recorded with multiple voices and other sounds, and the latter one extracts only the speech of the desired speaker. These technologies make speech more understandable for humans and contribute to improving downstream speech applications. Two important approaches are discussed: signal-model-based and neural-network-based methods. Then detailed explanations of representative techniques in the approaches, blind source separation in reverberant environments, and target speech extraction based on voice features are provided. Finally, the future prospects of this technological field are discussed.</p><p>複数の音声やそのほかの音が混ざって収録された音響信号から，個々の音を分離して抽出する音源分離，及び特定の話者の音声のみを抽出する目的音声抽出について，最新の技術動向を解説する．これらの技術は，人にとって音声をより聞き取りやすくするだけでなく，後段の音声アプリケーションの性能向上にも寄与する．二つの重要なアプローチとして，信号のモデルに基づく方法とニューラルネットワークに基づく方法をとりあげる．各アプローチの概要と特徴を述べ，代表的な技術として，残響環境下でのブラインド音源分離と声の特徴に基づく目的音声抽出について詳しく紹介する．最後に，この技術分野の今後の展望についても触れる．</p>
- DOI
- 10.1587/essfr.18.4_267
- Access Restrictions
- インターネット公開
- Data Provider (Database)
- 科学技術振興機構 : J-STAGE