Shinnosuke Takamichi 研究室

主宰者：Shinnosuke Takamichi

東京大学

AI 要約（直近 5 年の研究成果）

本研究室は、音声や音響現象を機械学習により理解・生成する研究を展開しています。具体的には、テキストから音声を合成する技術、環境音や音楽を生成する技術、複数の音源が混在する状況から目的の音声を抽出する技術など、多様な音響処理課題に取り組んでいます。また、人間のように自然な発話を生成するための音声品質評価や、音声と意味的な関連性を測る自動評価手法の開発にも注力しており、これらの技術を動画解説や人間とロボットの対話システムなど実世界の応用へ結びつけています。手法面では、深層学習モデル（特にトランスフォーマーやニューラルネットワーク）を活用し、大規模なデータから音響パターンを自動で学習する自己教師あり学習を重視しています。同時に、限られたデータから効率的に学習する工夫として、能動学習やデータ選別の最適化に取り組んでいます。さらに、音声を数値の列で表現するニューラル音響コーデックなど、音の本質的な特性をより良く捉える表現方法の研究も進めています。これらの研究を通じて、機械が人間らしい音声表現や多様な音響現象を理解・生成できる基盤技術の構築を目指すとともに、言語学的な法則（例えば自然言語に見られる統計的パターン）が合成音声にも現れるかといった基礎的な問いにも向き合っています。

※ AI（Claude）が、公開されている論文要旨から研究の問い・手法・主要な発見を事実情報として抽出・再構成して自動生成しています。誤りを含む可能性があるため、正確性は研究室公式情報でご確認ください。

外部リンク

研究成果（89 件）

[2026] Three-Stage BSRNN for Universal Speech Enhancement and Data Curation Using a Large Pre-Trained Speech Restoration Model
DOI: https://doi.org/10.1109/icassp55912.2026.11464365
[2026] J-SPAW2: A Japanese Corpus for Speaker Verification and Anti-Spoofing with Challenging Replay and Speech Synthesis Attacks
DOI: https://doi.org/10.21437/odyssey.2026-52
[2026] MangaVox: Dataset of Acted Voices Aligned with Manga Images Towards Computer Understanding of Audio Comics
DOI: https://doi.org/10.1109/icassp55912.2026.11463396
[2026] XACLE Challenge 2026: The First X-to-Audio Alignment Challenge
DOI: https://doi.org/10.1109/icassp55912.2026.11464875
[2026] Spatial-CLAP: Learning Spatially-Aware Audio–Text Embeddings for Multi-Source Conditions
DOI: https://doi.org/10.1109/icassp55912.2026.11460767
[2026] SS-JDSC: Single-Speaker Japanese Dysarthric Speech Corpus
DOI: https://doi.org/10.1109/icassp55912.2026.11464441
[2025] SpoofCeleb: Speech Deepfake Detection and SASV in the Wild
DOI: https://doi.org/10.1109/ojsp.2025.3529377
[2025] TTSOps: A Closed-Loop Corpus Optimization Framework for Training Multi-Speaker TTS Models From Dark Data
DOI: https://doi.org/10.1109/taslpro.2025.3633089
[2025] Toward Data-Efficient Speech Synthesis: Active Learning-Based Corpus Construction for Multi-Speaker Text-to-Speech Synthesis
DOI: https://doi.org/10.1109/access.2025.3645005
[2025] Excitement-Inducing Commentary Text-to-Speech System for Fighting Game Video Scenes
DOI: https://doi.org/10.1109/access.2025.3648378

続きを表示（残り 79 件）

[2025] Analysing the Language of Neural Audio Codecs
DOI: https://doi.org/10.1109/asru65441.2025.11434618
[2025] Learning Marmoset Vocal Patterns with a Masked Autoencoder for Robust Call Segmentation, Classification, and Caller Identification
DOI: https://doi.org/10.1109/asru65441.2025.11434811
[2025] Active Learning for Text-to-Speech Synthesis with Informative Sample Collection
DOI: https://doi.org/10.1109/apsipaasc65261.2025.11249057
[2025] Drum-to-Vocal Percussion Sound Conversion and Its Evaluation Methodology
DOI: https://doi.org/10.1109/apsipaasc65261.2025.11249004
[2025] Constructing an In-the-Wild Spoken Dialogue Dataset Based on Youtube Dialogue Videos
DOI: https://doi.org/10.1109/apsipaasc65261.2025.11249350
[2025] Do statistical patterns in neural audio codec tokens by synthesized speech reveal structure beyond speech quality?
DOI: https://doi.org/10.1121/10.0040578
[2025] Analysis of a dataset for evaluating semantic relevance between text and audio
DOI: https://doi.org/10.1121/10.0040371
[2025] Developing learners’ communicative competence in English as an international language using non-native English variations with AI speech synthesis technology
DOI: https://doi.org/10.4995/eurocall2025.2025.21210
[2025] RELATE: Subjective evaluation dataset for automatic evaluation of relevance between text and audio
DOI: https://doi.org/10.21437/interspeech.2025-1830
[2025] Measuring Time Delay Tolerance in Third-Person Live Commentary for Super Smash Bros. Ultimate
DOI: https://doi.org/10.1109/cog64752.2025.11114300
[2025] The Text-to-speech in the Wild (TITW) Database
DOI: https://doi.org/10.21437/interspeech.2025-2536
[2025] Exploring the Effect of Segmentation and Vocabulary Size on Speech Tokenization for Speech Language Models
DOI: https://doi.org/10.21437/interspeech.2025-310
[2025] J-SPAW: Japanese speaker verification and spoofing attacks recorded in-the-wild dataset
DOI: https://doi.org/10.21437/interspeech.2025-352
[2025] Voice Conversion for Likability Control via Automated Rating of Speech Synthesis Corpora
DOI: https://doi.org/10.21437/interspeech.2025-383
[2025] Language-queried target speech extraction using para-linguistic and non-linguistic prompts
DOI: https://doi.org/10.1250/ast.e25.27
[2024] JVNV: A Corpus of Japanese Emotional Speech With Verbal Content and Nonverbal Expressions
DOI: https://doi.org/10.1109/access.2024.3360885
[2024] DNN-Based Ensemble Singing Voice Synthesis With Interactions Between Singers
DOI: https://doi.org/10.1109/slt61566.2024.10832340
[2024] SpeechBERTScore: Reference-Aware Automatic Evaluation of Speech Generation Leveraging NLP Evaluation Metrics
DOI: https://doi.org/10.21437/interspeech.2024-1508
[2024] SaSLaW: Dialogue Speech Corpus with Audio-visual Egocentric Information Toward Environment-adaptive Dialogue Speech Synthesis
DOI: https://doi.org/10.21437/interspeech.2024-1554
[2024] Textless Dependency Parsing by Labeled Sequence Prediction
DOI: https://doi.org/10.21437/interspeech.2024-783
[2024] Noise-Robust Voice Conversion by Conditional Denoising Training Using Latent Variables of Recording Quality and Environment
DOI: https://doi.org/10.21437/interspeech.2024-972
[2024] Spatial Voice Conversion: Voice Conversion Preserving Spatial Information and Non-target Signals
DOI: https://doi.org/10.21437/interspeech.2024-1107
[2024] SRC4VC: Smartphone-Recorded Corpus for Voice Conversion Benchmark
DOI: https://doi.org/10.21437/interspeech.2024-388
[2024] Who Finds This Voice Attractive? A Large-Scale Experiment Using In-the-Wild Data
DOI: https://doi.org/10.21437/interspeech.2024-173
[2024] Do Learned Speech Symbols Follow Zipf’s Law?
DOI: https://doi.org/10.1109/icassp48485.2024.10448331
[2024] Real-Time Noise Estimation for Lombard-Effect Speech Synthesis in Human–Avatar Dialogue Systems
DOI: https://doi.org/10.1109/apsipaasc63619.2025.10848599
[2024] NecoBERT: Self-Supervised Learning Model Trained by Masked Language Modeling on Rich Acoustic Features Derived from Neural Audio Codec
DOI: https://doi.org/10.1109/apsipaasc63619.2025.10849090
[2024] Diversity-Based Core-Set Selection for Text-to-Speech with Linguistic and Acoustic Features
DOI: https://doi.org/10.1109/icassp48485.2024.10448068
[2024] Environmental Sound Synthesis from Vocal Imitations and Sound Event Labels
DOI: https://doi.org/10.1109/icassp48485.2024.10446906
[2024] Emotion-controllable Speech Synthesis Using Emotion Soft Label, Utterance-level Prosodic Factors, and Word-level Prominence
DOI: https://doi.org/10.1561/116.00000242
[2024] Text-Inductive Graphone-Based Language Adaptation for Low-Resource Speech Synthesis
DOI: https://doi.org/10.1109/taslp.2024.3369537
[2023] Visual Onoma-to-Wave: Environmental Sound Synthesis from Visual Onomatopoeias and Sound-Source Images
DOI: https://doi.org/10.1109/icassp49357.2023.10096517
[2023] Yodas: Youtube-Oriented Dataset for Audio and Speech
DOI: https://doi.org/10.1109/asru57964.2023.10389689
[2023] JNV corpus: A corpus of Japanese nonverbal vocalizations with diverse phrases and emotions
DOI: https://doi.org/10.1016/j.specom.2023.103004
[2023] Improving robustness of spontaneous speech synthesis with linguistic speech regularization and pseudo-filled-pause insertion
DOI: https://doi.org/10.21437/ssw.2023-10
[2023] Effects of text-to-speech synthesized speech on learners’ presentation anxiety and self-efficacy: A comparison of two models
DOI: https://doi.org/10.4995/eurocall2023.2023.16916
[2023] ChatGPT-EDSS: Empathetic Dialogue Speech Synthesis Trained from ChatGPT-derived Context Word Embeddings
DOI: https://doi.org/10.21437/interspeech.2023-1095
[2023] Laughter Synthesis using Pseudo Phonetic Tokens with a Large-scale In-the-wild Laughter Corpus
DOI: https://doi.org/10.21437/interspeech.2023-806
[2023] How Generative Spoken Language Modeling Encodes Noisy Speech: Investigation from Phonetics to Syntactics
DOI: https://doi.org/10.21437/interspeech.2023-981
[2023] CALLS: Japanese Empathetic Dialogue Speech Corpus of Complaint Handling and Attentive Listening in Customer Center
DOI: https://doi.org/10.21437/interspeech.2023-1098
[2023] HumanDiffusion: diffusion model using perceptual gradients
DOI: https://doi.org/10.21437/interspeech.2023-1680
[2023] COCO-NUT: Corpus of Japanese Utterance and Voice Characteristics Description for Prompt-Based Control
DOI: https://doi.org/10.1109/asru57964.2023.10389693
[2023] Learning to Speak from Text: Zero-Shot Multilingual Text-to-Speech with Unsupervised Text Pretraining
DOI: https://doi.org/10.24963/ijcai.2023/575
[2023] Improving Speech Prosody of Audiobook Text-To-Speech Synthesis with Acoustic and Textual Contexts
DOI: https://doi.org/10.1109/icassp49357.2023.10096247
[2023] TEXT-TO-SPEECH SYNTHESIS FROM DARK DATA WITH EVALUATION-IN-THE-LOOP DATA SELECTION
DOI: https://doi.org/10.1109/icassp49357.2023.10095161
[2023] jaCappella Corpus: A Japanese a Cappella Vocal Ensemble Corpus
DOI: https://doi.org/10.1109/icassp49357.2023.10095569
[2023] MID-Attribute Speaker Generation Using Optimal-Transport-Based Interpolation of Gaussian Mixture Models
DOI: https://doi.org/10.1109/icassp49357.2023.10097113
[2023] TimToShape: Supporting Practice of Musical Instruments by Visualizing Timbre with 2D Shapes based on Crossmodal Correspondences
DOI: https://doi.org/10.1145/3581641.3584053
[2023] VTTS: Visual-Text To Speech
DOI: https://doi.org/10.1109/slt54892.2023.10022739
[2023] SelfRemaster: Self-Supervised Speech Restoration for Historical Audio Resources
DOI: https://doi.org/10.1109/access.2023.3345027
[2022] Empirical Study Incorporating Linguistic Knowledge on Filled Pauses for Personalized Spontaneous Speech Synthesis
DOI: https://doi.org/10.23919/apsipaasc55919.2022.9979895
[2022] Development of a hands‐free electrolarynx for obtaining a human‐like voice using the LPC residual wave
DOI: https://doi.org/10.1002/eej.23409
[2022] SelfRemaster: Self-Supervised Speech Restoration with Analysis-by-Synthesis Approach Using Channel Modeling
DOI: https://doi.org/10.21437/interspeech.2022-298
[2022] Acoustic Modeling for End-to-End Empathetic Dialogue Speech Synthesis Using Linguistic and Prosodic Contexts of Dialogue History
DOI: https://doi.org/10.21437/interspeech.2022-403
[2022] STUDIES: Corpus of Japanese Empathetic Dialogue Speech Towards Friendly Voice Agent
DOI: https://doi.org/10.21437/interspeech.2022-300
[2022] J-MAC: Japanese multi-speaker audiobook corpus for speech synthesis
DOI: https://doi.org/10.21437/interspeech.2022-444
[2022] Predicting VQVAE-based Character Acting Style from Quotation-Annotated Text for Audiobook Speech Synthesis
DOI: https://doi.org/10.21437/interspeech.2022-638
[2022] Development of a Hands-free Electrolarynx for Obtaining a Human-like Voice using the LPC Residual Wave
DOI: https://doi.org/10.1541/ieejfms.142.390
[2022] How Should We Evaluate Synthesized Environmental Sounds
DOI: https://doi.org/10.23919/apsipaasc55919.2022.9980187
[2022] Robustness of Signal Processing-Based Pseudonymization Method Against Decryption Attack
DOI: https://doi.org/10.21437/odyssey.2022-40
[2022] Personalized Filled-pause Generation with Group-wise Prediction Models
DOI: https://doi.org/10.63317/5esa64wph3ou
[2022] UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022
DOI: https://doi.org/10.21437/interspeech.2022-439
[2022] Onoma-to-wave: Environmental Sound Synthesis from Onomatopoeic Words
DOI: https://doi.org/10.1561/116.00000049
[2022] Hands-Free Wearable Electrolarynx using Linear Predictive Coding Residual Waves and Listening Evaluation
DOI: https://doi.org/10.14326/abe.11.68
[2021] Perceptual-Similarity-Aware Deep Speaker Representation Learning for Multi-Speaker Generative Modeling
DOI: https://doi.org/10.1109/taslp.2021.3059114
[2021] Lightweight Voice Anonymization Based on Data-Driven Optimization of Cascaded Voice Modification Modules
DOI: https://doi.org/10.1109/slt48900.2021.9383535
[2021] Low-Latency Incremental Text-to-Speech Synthesis with Distilled Context Prediction Network
DOI: https://doi.org/10.1109/asru51503.2021.9687904
[2021] Lightweight and irreversible speech pseudonymization based on data-driven optimization of cascaded voice modification modules
DOI: https://doi.org/10.1016/j.csl.2021.101315
[2021] DNN-Based Low-Musical-Noise Single-Channel Speech Enhancement Based on Higher-Order-Moments Matching
DOI: https://doi.org/10.1587/transinf.2021edp7041
[2021] Digital Speech Makeup: Voice Conversion Based Altered Auditory Feedback for Transforming Self-Representation
DOI: https://doi.org/10.1145/3462244.3479934
[2021] Cross-Lingual Speaker Adaptation Using Domain Adaptation and Speaker Consistency Loss for Text-To-Speech Synthesis
DOI: https://doi.org/10.21437/interspeech.2021-897
[2021] Audiobook Speech Synthesis Conditioned by Cross-Sentence Context-Aware Word Embeddings
DOI: https://doi.org/10.21437/ssw.2021-37
[2021] Accent Modeling of Low-Resourced Dialect in Pitch Accent Language Using Variational Autoencoder
DOI: https://doi.org/10.21437/ssw.2021-33
[2021] Real-Time Full-Band Voice Conversion with Sub-Band Modeling and Data-Driven Phase Estimation of Spectral Differentials
DOI: https://doi.org/10.1587/transinf.2020edp7252
[2021] Disentangled Speaker and Language Representations Using Mutual Information Minimization and Domain Adaptation for Cross-Lingual TTS
DOI: https://doi.org/10.1109/icassp39728.2021.9414226
[2021] Humanacgan: Conditional Generative Adversarial Network with Human-Based Auxiliary Classifier and its Evaluation in Phoneme Perception
DOI: https://doi.org/10.1109/icassp39728.2021.9413363
[2021] Noise Robust Acoustic Anomaly Detection System with Nonnegative Matrix Factorization Based on Generalized Gaussian Distribution
DOI: https://doi.org/10.1587/transinf.2020edk0002
[2021] End-to-end incremental TTS with lookahead generation with large pretrained language model
[2021] Fundamentals, Recent Trends, and Future Direction of Voice Conversion
DOI: https://doi.org/10.1541/ieejjournal.141.93

科研費（0 件）

まだデータがありません（KAKEN 取り込み後に表示）。

所属学会・役職（0 件）

まだデータがありません（学会データ連携後に表示）。

AI 要約（直近 5 年の研究成果）

外部リンク

関連研究室(8 件)

研究成果（89 件）

科研費（0 件）

所属学会・役職（0 件）