Publications

Speech generation for indigenous language education
Cite DOI URL
Improving curriculum learning for target speaker extraction with synthetic speakers
Cite URL
Disentangling the Prosody and Semantic Information with Pre-trained Model for In-Context Learning based Zero-Shot Voice Conversion
Cite URL
The VoiceMOS Challenge 2024: Beyond Speech Quality Prediction
Cite URL
Disentangling the Prosody and Semantic Information with Pre-trained Model for In-Context Learning based Zero-Shot Voice Conversion
Cite URL
Spoofing-Aware Speaker Verification Robust Against Domain and Channel Mismatches
Cite URL
Quantifying Source Speaker Leakage in One-to-One Voice Conversion
Cite arXiv
A Preliminary Case Study on Long-Form In-the-Wild Audio Spoofing Detection
Cite arXiv
Malacopula: adversarial automatic speaker verification attacks using a neural-based generalised Hammerstein model
Cite arXiv
Exploring Active Data Selection Strategies for Continuous Training in Deepfake Detection
Cite URL
ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale
Cite DOI URL
Polarity Classification of Low Resource Roman Urdu and Movie Reviews Sentiments Using Machine Learning-Based Ensemble Approaches
Cite Preprint
The VoicePrivacy 2022 Challenge: Progress and Perspectives in Voice Anonymisation
Cite DOI URL
Objective assessment of synthetic speech and the VoiceMOS Challenge (合成音声の客観評価とVoiceMOSチャレンジ)
Cite DOI URL
Revisiting and Improving Scoring Fusion for Spoofing-aware Speaker Verification Using Compositional Data Analysis
Cite URL
Generating Speakers by Prompting Listener Impressions for Pre-trained Multi-Speaker Text-to-Speech Systems
Cite URL
An Initial Investigation of Language Adaptation for TTS Systems under Low-resource Scenarios
Cite URL
Target Speaker Extraction with Curriculum Learning
Cite URL
Spoof Diarization:" What Spoofed When" in Partially Spoofed Audio
Cite URL
To what extent can ASV systems naturally defend against spoofing attacks?
Cite URL
Uncertainty as a Predictor: Leveraging Self-Supervised Learning for Zero-Shot MOS Prediction
Cite URL
SynVox2: Towards a privacy-friendly VoxCeleb2 dataset
Cite Preprint
SPOOFING ATTACK AUGMENTATION: CAN DIFFERENTLY-TRAINED ATTACK MODELS IMPROVE GENERALISATION?
Cite URL
Detection of Sarcasm in Urdu Tweets using Deep Learning and Transformer based Hybrid Approaches
Cite URL
DDSP-based Neural Waveform Synthesis of Polyphonic Guitar Performance from String-wise MIDI Input
Cite Preprint Demo
Can large-scale vocoded spoofed data improve speech spoofing countermeasure with a self-supervised front end?
Cite URL
eKYC-DF: A Large-Scale Deepfake Dataset for Developing and Evaluating eKYC Systems
Cite DOI URL
Bridging Textual and Tabular Worlds for Fact Verification: A Lightweight, Attention-Based Model
Cite URL Preprint
Analysis of Fine-Grained Counting Methods for Masked Face Counting: A Comparative Study
Cite DOI URL
A review on subjective and objective evaluation of synthetic speech
Cite DOI URL
The VoiceMOS Challenge 2023:Zero-shot Subjective Speech Quality Prediction for Multiple Domains
Cite URL
Partial Rank Similarity Minimization Method for Quality MOS Prediction of Unseen Speech Synthesis Systems in Zero-Shot and Semi-supervised setting
Cite URL
XFEVER: Exploring Fact Verification across Languages
Cite URL
Speaker Anonymization Using Orthogonal Householder Neural Network
Cite DOI Preprint
How Close are Other Computer Vision Tasks to Deepfake Detection?
Cite
Towards Single Integrated Spoofing-aware Speaker Verification Embeddings
Cite DOI
SASPEECH: A Hebrew Single Speaker Dataset for Text To Speech and Voice Conversion
Cite DOI Database
Range-Based Equal Error Rate for Spoof Localization
Cite DOI Preprint
Investigating Range-Equalizing Bias in Mean Opinion Score Ratings of Synthesized Speech
Cite DOI Preprint
Improving Generalization Ability of Countermeasures for New Mismatch Scenario by Combining Multiple Advanced Regularization Terms
Cite DOI Preprint Code
Controlling Multi-Class Human Vocalization Generation via a Simple Segment-based Labeling Scheme
Cite DOI
Exploring Isolated Musical Notes as Pre-training Data for Predominant Instrument Recognition in Polyphonic Music
Cite arXiv
BodyFormer: Semantics-Guided 3D Body Gesture Synthesis with Transformer
Automatic gesture synthesis from speech is a topic that has attracted researchers for applications in remote communication, video games …
Cite DOI URL
Revisiting Pathologies of Neural Models under Input Reduction
Cite URL Code Checkpoints
ASVspoof 2021: Towards Spoofed and Deepfake Speech Detection in the Wild
Cite DOI Preprint Website
Xiaoicesing 2: A High-Fidelity Singing Voice Synthesizer Based on Generative Adversarial Network
Cite Preprint
Spoofed Training Data for Speech Spoofing Countermeasure Can Be Efficiently Created Using Neural Vocoders
Cite DOI Preprint
Joint Noise Reduction and Listening Enhancement for Full-End Speech Enhancement
Cite DOI Preprint
Hiding Speaker’s Sex in Speech Using Zero-Evidence Speaker Representation in an Analysis/Synthesis Pipeline
Cite DOI Preprint Codes and samples Pre-trained models
Can Knowledge of End-to-End Text-to-Speech Models Improve Neural MIDI-to-Audio Synthesis Systems?
Cite DOI Preprint Samples Pre-trained models
Analysis of Master Vein Attacks on Finger Vein Recognition Systems
Cite DOI
ZMM-TTS: Zero-shot Multilingual and Multispeaker Speech Synthesis Conditioned on Self-supervised Discrete Speech Representations
Cite URL
The PartialSpoof Database and Countermeasures for the Detection of Short Fake Speech Segments Embedded in an Utterance
Automatic speaker verification is susceptible to various manipulations and spoofing, such as text-to-speech synthesis, voice …
Cite DOI Preprint video
SSI-Net: A Multi-Stage Speech Signal Improvement System for ICASSP 2023 SSI Challenge
Cite DOI
Speaker-Text Retrieval via Contrastive Learning
Cite arXiv
Investigating Active-Learning-Based Training Data Selection for Speech Spoofing Countermeasure
Cite DOI Preprint
Cyber Vaccine for Deepfake Immunity
Cite DOI Preprint
CrossSinger: A Cross-Lingual Multi-Singer High-Fidelity Singing Voice Synthesizer Trained on Monolingual Singers
Cite arXiv
Outlier-Aware Training for Improving Group Accuracy Disparities
Methods addressing spurious correlations such as Just Train Twice (JTT, Liu et al. 2021) involve reweighting a subset of the training …
PDF Cite Preprint Codes Pre-trained models
Use of Speaker Recognition Approaches for Learning and Evaluating Embedding Representations of Musical Instrument Sounds
Cite DOI
The VoicePrivacy 2020 Challenge: Results and findings
Cite Paper Preprint challenge website
The VoiceMOS Challenge 2022
Cite DOI Preprint CodaLab website
SVSNet: An End-to-End Speaker Voice Similarity Assessment Model
Cite DOI Preprint
Spoofing-Aware Attention based ASV Back-end with Multiple Enrollment Utterances and a Sampling Strategy for the SASV Challenge 2022
Cite DOI Preprint
Robust Deepfake on Unrestricted Media: Generation and Detection
Recent advances in deep learning have led to substantial improvements in deepfake generation, resulting in fake media with a more …
PDF Cite DOI
Optimizing Tandem Speaker Verification and Anti-Spoofing Systems
Cite DOI
On the Interplay between Sparsity, Naturalness, Intelligibility, and Prosody in Speech Synthesis
Cite DOI Preprint Samples
Mitigating the Diminishing Effect of Elastic Weight Consolidation
Cite PDF Codes
Master Face Attacks on Face Recognition Systems
Cite DOI
Language-Independent Speaker Anonymization Approach Using Self-Supervised Pre-Trained Models
Cite DOI Preprint Samples Codes
Investigating Self-Supervised Front Ends for Speech Spoofing Countermeasures
Cite DOI Preprint
Improving Neural-Network-Based Speech Enhancement for Noise Reduction and Intelligibility Boosting(雑音抑圧と明瞭性向上のためのニューラルネットワークによる音声強調の改善について)
Cite URL
Generalization Ability of MOS Prediction Networks
Cite DOI Preprint Codes
Future Trends in Digital Face Manipulation and Detection
Recently, digital face manipulationDigital face manipulation and its detection have sparked large interest in industry and academia …
PDF Cite DOI
Estimating the Confidence of Speech Spoofing Countermeasure
Cite DOI Preprint
DDS: A new device-degraded speech dataset for speech enhancement
Cite DOI Preprint Database DDS database (DAPS portion) DDS database (VCTK portion part 1) DDS database (VCTK portion part 2)
Capsule-Forensics Networks for Deepfake Detection
SeveralCapsule-Forensics sophisticated convolutional neural network (CNN)Convolutional Neural Networks (CNN) architectures have been …
PDF Cite DOI
Automatic Speaker Verification Spoofing and Deepfake Detection Using Wav2vec 2.0 and Data Augmentation
Cite DOI Preprint
Attention Back-End for Automatic Speaker Verification with Multiple Enrollment Utterances
Cite DOI Preprint Codes
Analyzing Language-Independent Speaker Anonymization Framework under Unseen Conditions
Cite DOI Preprint Samples
A Practical Guide to Logical Access Voice Presentation Attack Detection
Voice-based human-machine interfaces with an automatic speaker verification (ASV) component are commonly used in the market. However, …
PDF Cite DOI
A Multi-Level Attention Model for Evidence-Based Fact Checking
Cite DOI URL
Text-to-Speech Synthesis Techniques for MIDI-to-Audio Synthesis
Cite DOI Preprint Samples
Revisiting Speech Content Privacy
Cite DOI Preprint
Rakugo Speech Synthesis: Toward Speech Synthesis That Entertains Audiences(落語音声合成: 人を楽しませる音声合成に向けて)
Cite URL
OpenForensics: Large-Scale Challenging Dataset For Multi-Face Forgery Detection And Segmentation In-The-Wild
Cite DOI Preprint Database
Multi-task Learning in Utterance-level and Segmental-level Spoof Detection
Cite DOI Preprint Database
Multi-Metric Optimization Using Generative Adversarial Networks for Near-End Speech Intelligibility Enhancement
Cite DOI Preprint Samples Codes
Lexical pitch accent and duration modeling for neural end-to-end text-to-speech synthesis(End-to-endニューラル音声合成のための構文的ピッチアクセントと音素継続長のモデル化)
Cite URL
Learning Disentangled Phone and Speaker Representations in a Semi-Supervised VQ-VAE Paradigm
Cite DOI Preprint Samples
How Similar or Different is Rakugo Speech Synthesizer to Professional Performers?
Cite DOI Preprint
How do Voices from Past Speech Synthesis Challenges Compare Today?
Cite DOI Preprint
Exploring Disentanglement with Multilingual and Monolingual VQ-VAE
Cite DOI Preprint Samples
Enhancing Low-Quality Voice Recordings Using Disentangled Channel Factor and Neural Waveform Model
Cite DOI Preprint Samples
End-to-End Text-to-Speech Using Latent Duration Based on VQ-VAE
Cite DOI Preprint Samples
Effectiveness of Detection-based and Regression-based Approaches for Estimating Mask-Wearing Ratio
Cite DOI Preprint
Denoising-and-Dereverberation Hierarchical Neural Vocoder for Robust Waveform Generation
Cite DOI Preprint Samples
Benchmarking and challenges in security and privacy for voice biometrics
Cite DOI Preprint
ASVspoof 2021: accelerating progress in spoofed and deepfake speech detection
Cite DOI Preprint challenge website
ASVspoof 2019: Spoofing Countermeasures for the Detection of Synthesized, Converted and Replayed Speech
Cite DOI Preprint
An Initial Investigation for Detecting Partially Spoofed Audio
Cite DOI Preprint Samples Database
A Comparative Study on Recent Neural Spoofing Countermeasures for Synthetic Speech Detection
Cite DOI Preprint Codes
Deep learning based voice cloning framework for a unified system of text-to-speech and voice conversion（テキスト音声合成と声質変換を統合した深層学習によるボイスクローニング）
Cite URL
Introduction to Voice Presentation Attack Detection and Recent Advances
Over the past few years, significant progress has been made in the field of presentation attack detection (PAD) for automatic speaker …
PDF Cite DOI
Fundamental Frequency Modeling for Neural-Network-Based Statistical Parametric Speech Synthesis (ニューラルネットワークに基づく統計的音声合成のための基本周波数モデリング)
Cite URL
User Generated Dialogue Systems: uDialogue
This chapter introduces the idea of user-generated dialogueUDialogue technologycontent and describes our experimental exploration aimed …
PDF Cite DOI
おしゃべりなコンピュータ : 音声合成技術の現在と未来
PDF Cite
Speaker Recognition Anti-spoofing
Progress in the development of spoofing countermeasures for automatic speaker recognition is less advanced than equivalent work …
PDF Cite DOI
Building personalized synthetic voices for individuals with dysarthria using the HTS toolkit
Cite
A review on subjective and objective evaluation of synthetic speech
DOI URL