InterSpeech 2021

Mapa webu

• Domů

• InterSpeech 2021
  • Keynotes (4)
    • ISCA Medalist: Forty years of speech and language processing: from Bayes decision rule to deep learning
    • Ethical and Technological Challenges of Conversational AI
    • Adaptive listening to everyday soundscapes
    • Language Modeling and Artificial Intelligence
  • Survey talks (4)
    • Towards automatic speech recognition for people with atypical speech
    • Uncovering the acoustic cues of COVID-19 infection
    • Learning speech models from multi-modal data
    • Child Language Acquisition studied with Wearables
  • Acoustic event detection and acoustic scene classification (5)
    • SpecMix : A Mixed Sample Data Augmentation method for Training with Time-Frequency Domain Features
      (3 minutes introduction)
    • Acoustic Scene Classification using Kervolution-Based SubSpectralNet
      (3 minutes introduction)
    • Event Specific Attention for Polyphonic Sound Event Detection
      (3 minutes introduction)
    • AST: Audio Spectrogram Transformer
      (3 minutes introduction)
    • Shallow Convolution-Augmented Transformer with Differentiable Neural Computer for Low-Complexity Classification of Variable-Length Acoustic Scene
      (3 minutes introduction)
  • Applications in transcription, education and learning (8)
    • Weakly-supervised word-level pronunciation error detection in non-native English speech
      (longer introduction)
    • End-to-End Speaker-Attributed ASR with Transformer
      (3 minutes introduction)
    • Explore Wav2vec 2.0 for Mispronunciation Detection
      (3 minutes introduction)
    • Lexical Density Analysis of Word Productions in Japanese English Using Acoustic Word Embeddings
      (3 minutes introduction)
    • Deep feature transfer learning for automatic pronunciation assessment
      (3 minutes introduction)
    • "You don't understand me!": Comparing ASR results for L1 and L2 speakers of Swedish
      (3 minutes introduction)
    • NeMo Inverse Text Normalization: From Development To Production
      (3 minutes introduction)
    • Improvement of Automatic English Pronunciation Assessment with Small Number of Utterances Using Sentence Speakability
      (3 minutes introduction)
  • ASR Technologies and systems (1)
    • Fearless Steps Challenge Phase-3 (FSC P3): Advancing SLT for Unseen Channel and Mission data across NASA Apollo audio
      (Oral presentation)
  • Assessment of pathological speech and language I (4)
    • Automatic extraction of speech rhythm descriptors for speech intelligibility assessment in the context of head and neck cancers
      (Oral presentation)
    • Speech Disorder Classification Using Extended Factorized Hierarchical Variational Auto-encoders
      (Oral presentation)
    • The Impact of Forced-Alignment Errors on Automatic Pronunciation Evaluation
      (Oral presentation)
    • Neural Speaker Embeddings for Ultrasound-based Silent Speech Interfaces
      (Oral presentation)
  • Assessment of pathological speech and language II (13)
    • Speech intelligibility of dysarthric speech: human scores and acoustic-phonetic features
      (3 minutes introduction)
    • Analyzing short term dynamic speech features for understanding behavioral traits of children with autism spectrum disorder
      (3 minutes introduction)
    • Analyzing short term dynamic speech features for understanding behavioral traits of children with autism spectrum disorder
      (longer introduction)
    • Vocalization Recognition of People with Profound Intellectual and Multiple Disabilities (PIMD) Using Machine Learning Algorithms
      (3 minutes introduction)
    • Detection of Consonant Errors in Disordered Speech Based on Consonant-vowel Segment Embedding
      (3 minutes introduction)
    • Detection of Consonant Errors in Disordered Speech Based on Consonant-vowel Segment Embedding
      (longer introduction)
    • Assessing Posterior-Based Mispronunciation Detection on Field-Collected Recordings from Child Speech Therapy Sessions
      (3 minutes introduction)
    • Identifying cognitive impairment using sentence representation vectors
      (3 minutes introduction)
    • Parental spoken scaffolding and narrative skills in crowd-sourced storytelling samples of young children
      (3 minutes introduction)
    • Uncertainty-Aware COVID-19 Detection from Imbalanced Sound Data
      (3 minutes introduction)
    • Source and Vocal Tract Cues for Speech-based Classification of Patients with Parkinson’s Disease and Healthy Subjects
      (3 minutes introduction)
    • Source and Vocal Tract Cues for Speech-based Classification of Patients with Parkinson’s Disease and Healthy Subjects
      (longer introduction)
    • CLAC: A Speech Corpus Of Healthy English Speakers
      (3 minutes introduction)
  • Automatic Speech Recognition in Air Traffic Management (4)
    • Detecting English Speech in the Air Traffic Control Voice Communication
      (Oral presentation)
    • Contextual Semi-Supervised Learning: An Approach To Leverage Air-Surveillance and Untranscribed ATC Data in ASR Systems
      (Oral presentation)
    • Boosting of contextual information in ASR for air-traffic call-sign recognition
      (Oral presentation)
    • Modeling the effect of military oxygen masks on speech characteristics
      (Oral presentation)
  • Communication and interaction, multimodality (8)
    • A Psychology-Driven Computational Analysis of Political Interviews
      (3 minutes introduction)
    • Speech Emotion Recognition based on Attention Weight Correction Using Word-level Confidence Measure
      (3 minutes introduction)
    • Speech Emotion Recognition based on Attention Weight Correction Using Word-level Confidence Measure
      (longer introduction)
    • Effects of voice type and task on L2 learners’ awareness of pronunciation errors
      (3 minutes introduction)
    • Lexical Entrainment and Intra-Speaker Variability in Cooperative Dialogues
      (3 minutes introduction)
    • Detecting Alzheimer's Disease using Interactional and Acoustic features from spontaneous speech
      (3 minutes introduction)
    • Investigating the interplay between affective, phonatory and motoric subsystems in Autism Spectrum Disorder using an audiovisual dialogue agent
      (3 minutes introduction)
    • Analysis of eye gaze reasons and gaze aversions during three-party conversations
      (3 minutes introduction)
  • ConferencingSpeech 2021 challenge: Far-field Multi-Channel Speech Enhancement for Video Conferencing (5)
    • A Causal U-net based Neural Beamforming Network for Real-Time Multi-Channel Speech Enhancement
      (Oral presentation)
    • A Partitioned-Block Frequency-Domain Adaptive Kalman Filter for Stereophonic Acoustic Echo Cancellation
      (Oral presentation)
    • Real-Time Independent Vector Analysis Using Semi-Supervised Nonnegative Matrix Factorization as a Source Model
      (Oral presentation)
    • Improving Channel Decorrelation for Multi-Channel Target Speech Extraction
      (Oral presentation)
    • Inplace Gated Convolutional Recurrent Neural Network For Dual-channel Speech Enhancement
      (Oral presentation)
  • Cross/multi-lingual and code-switched ASR (7)
    • Bootstrap an End-to-end ASR System by Multilingual Training, Transfer Learning, Text-to-text Mapping and Synthetic Audio
      (3 minutes introduction)
    • Unsupervised Cross-lingual Representation Learning for Speech Recognition
      (3 minutes introduction)
    • Multilingual and code-switching ASR challenges for low resource Indian languages
      (3 minutes introduction)
    • SRI-B End-to-End System for Multilingual and Code-Switching ASR Challenges for Low Resource Indian Languages
      (3 minutes introduction)
    • Hierarchical Phone Recognition with Compositional Phonetics
      (3 minutes introduction)
    • Towards One Model to Rule All: Multilingual Strategy for Dialectal Code-Switching Arabic ASR
      (3 minutes introduction)
    • Differentiable Allophone Graphs for Language Universal Speech Recognition
      (3 minutes introduction)
  • Disordered speech (3)
    • Acoustic Indicators of Speech Motor Coordination in Adults With and Without Traumatic Brain Injury
      (Oral presentation)
    • On Modeling Glottal Source Information for Phonation Assessment in Parkinson’s Disease
      (Oral presentation)
    • EasyCall corpus: a dysarthric speech dataset
      (Oral presentation)
  • Diverse modes of speech acquisition and processing (10)
    • Segment and Tone Production in Continuous Speech of Hearing and Hearing-impaired Children
      (3 minutes introduction)
    • Effect of Carrier Bandwidth on Understanding Mandarin Sentences in Simulated Electric-acoustic Hearing
      (3 minutes introduction)
    • Effect of Carrier Bandwidth on Understanding Mandarin Sentences in Simulated Electric-acoustic Hearing
      (longer introduction)
    • A Comparative Study Of Different EMG Features For Acoustic-to-EMG Mapping
      (3 minutes introduction)
    • A Comparative Study Of Different EMG Features For Acoustic-to-EMG Mapping
      (longer introduction)
    • An Attention Self-supervised Contrastive Learning based Three-stage Model for Hand Shape Feature Representation in Cued Speech
      (3 minutes introduction)
    • An Attention Self-supervised Contrastive Learning based Three-stage Model for Hand Shape Feature Representation in Cued Speech
      (longer introduction)
    • Remote smartphone-based speech collection: acceptance and barriers in individuals with major depressive disorder
      (3 minutes introduction)
    • Remote smartphone-based speech collection: acceptance and barriers in individuals with major depressive disorder
      (longer introduction)
    • Silent versus modal multi-speaker speech recognition from ultrasound and video
      (3 minutes introduction)
  • Embedding and Network Architecture for Speaker Recognition (3)
    • A Thousand Words are Worth More Than One Recording: Word-Embedding Based Speaker Change Detection
      (Oral presentation)
    • Leveraging speaker attribute information using multi task learning for speaker verification and diarization
      (Oral presentation)
    • Spine2Net: SpineNet with Res2Net and Time-Squeeze-and-Excitation Blocks for Speaker Recognition
      (Oral presentation)
  • Emotion and Sentiment Analysis I (2)
    • Separation of Emotional and Reconstruction Embeddings on Ladder Network to Improve Speech Emotion Recognition Robustness in Noisy Conditions
      (Oral presentation)
    • M3: MultiModal Masking applied to sentiment analysis
      (Oral presentation)
  • Emotion and Sentiment Analysis II (9)
    • Automatic Analysis of the Emotional Content of Speech in Daylong Child-Centered Recordings from a Neonatal Intensive Care Unit
      (3 minutes introduction)
    • Multimodal Sentiment Analysis with Temporal Modality Attention
      (3 minutes introduction)
    • Stochastic Process Regression for Cross-Cultural Speech Emotion Recognition
      (3 minutes introduction)
    • Stochastic Process Regression for Cross-Cultural Speech Emotion Recognition
      (longer introduction)
    • Emotion Recognition from Speech Using Wav2vec 2.0 Embeddings
      (3 minutes introduction)
    • Applying TDNN Architectures for Analyzing Duration Dependencies on Speech Emotion Recognition
      (3 minutes introduction)
    • Applying TDNN Architectures for Analyzing Duration Dependencies on Speech Emotion Recognition
      (longer introduction)
    • Acoustic Features and Neural Representations for Categorical Emotion Recognition from Speech
      (3 minutes introduction)
    • Leveraging Pre-trained Language Model for Speech Sentiment Analysis
      (3 minutes introduction)
  • Emotion and Sentiment Analysis III (4)
    • Audio-Visual Speech Emotion Recognition by Disentangling Emotion and Identity Attributes
      (3 minutes introduction)
    • Metric Learning Based Feature Representation With Gated Fusion Model For Speech Emotion Recognition
      (3 minutes introduction)
    • Speech Emotion Recognition with Multi-task Learning
      (3 minutes introduction)
    • Generalized Dilated CNN Models for Depression Detection Using Inverted Vocal Tract Variables
      (3 minutes introduction)
  • Feature, Embedding and Neural Architecture for Speaker Recognition (8)
    • Bidirectional Multiscale Feature Aggregation for Speaker Verification
      (3 minutes introduction)
    • Improving Time Delay Neural Network Based Speaker Recognition With Convolutional Block And Feature Aggregation Methods
      (3 minutes introduction)
    • Improving Time Delay Neural Network Based Speaker Recognition With Convolutional Block And Feature Aggregation Methods
      (longer introduction)
    • Binary Neural Network for Speaker Verification
      (3 minutes introduction)
    • Mutual Information Enhanced Training for Speaker Embedding
      (3 minutes introduction)
    • Y-Vector: Multiscale Waveform Encoder for Speaker Embedding
      (3 minutes introduction)
    • Phoneme-aware and Channel-wise Attentive Learning for Text Dependent Speaker Verification
      (3 minutes introduction)
    • Serialized Multi-Layer Multi-Head Attention for Neural Speaker Embedding
      (3 minutes introduction)
  • Graph and End-to-End Learning for Speaker Recognition (1)
    • Reformulating DOVER-Lap Label Mapping as a Graph Partitioning Problem
      (Oral presentation)
  • Health and Affect I (3)
    • Measuring Voice Quality Parameters after Speaker Pseudonymization
      (Oral presentation)
    • Speaking Corona? Human and Machine Recognition of COVID-19 from Voice
      (Oral presentation)
    • Acoustic-Prosodic, Lexical and Demographic Cues to Persuasiveness in Competitive Debate Speeches
      (Oral presentation)
  • Health and Affect II (9)
    • Automatic Speech Recognition systems errors for objective sleepiness detection through voice
      (3 minutes introduction)
    • Robust Laughter Detection in Noisy Environments
      (3 minutes introduction)
    • Impact of Emotional State on Estimation of Willingness to Buy from Advertising Speech
      (3 minutes introduction)
    • Stacked Recurrent Neural Networks for Speech-Based Inference of Attachment Condition in School Age Children
      (3 minutes introduction)
    • Emotion Carrier Recognition from Personal Narratives
      (3 minutes introduction)
    • Non-verbal Vocalisation and Laughter Detection using Sequence-to-sequence Models and Multi-label Training
      (3 minutes introduction)
    • Visual Speech for Obstructive Sleep Apnea Detection
      (3 minutes introduction)
    • Analysis of Contextual Voice Changes in Remote Meetings
      (3 minutes introduction)
    • Speech based Depression Severity Level Classification Using a Multi-Stage Dilated CNN-LSTM Model
      (3 minutes introduction)
  • INTERSPEECH 2021 Acoustic Echo Cancellation Challenge (3)
    • INTERSPEECH 2021 Acoustic Echo Cancellation Challenge
      (Oral presentation)
    • F-T-LSTM based Complex Network for Joint Acoustic Echo Cancellation and Speech Enhancement
      (Oral presentation)
    • Acoustic Echo Cancellation using Deep Complex Neural Network with Nonlinear Magnitude Compression and Phase Information
      (Oral presentation)
  • INTERSPEECH 2021 Deep Noise Suppression Challenge (2)
    • Deep Noise Suppression With Non-Intrusive PESQNet Supervision Enabling the Use of Real Training Data
      (Oral presentation)
    • DBNet: A Dual-branch Network Architecture Processing on Spectrum and Waveform for Single-channel Speech Enhancement
      (Oral presentation)
  • Keyword search and spoken language processing (3)
    • Semantic sentence similarity: size does not always matter
      (Oral presentation)
    • Spoken Term Detection and Relevance Score Estimation using Dot-Product of Pronunciation Embeddings
      (Oral presentation)
    • Toward Genre Adapted Close Captioning
      (Oral presentation)
  • Language and Accent Recognition (3)
    • End-to-End Language Diarization for Bilingual Code-switching Speech
      (3 minutes introduction)
    • Exploring wav2vec 2.0 on speaker verification and language identification
      (3 minutes introduction)
    • E2E-based Multi-task Learning Approach to Joint Speech and Accent Recognition
      (3 minutes introduction)
  • Language and Lexical Modeling for ASR (8)
    • Incorporating External POS Tagger for Punctuation Restoration
      (3 minutes introduction)
    • Phonetically Induced Subwords for End-to-End Speech Recognition
      (3 minutes introduction)
    • Lookup-Table Recurrent Language Models for Long Tail Speech Recognition
      (3 minutes introduction)
    • Contextual Density Ratio for Language Model Biasing of Sequence to Sequence ASR Systems
      (3 minutes introduction)
    • Contextual Density Ratio for Language Model Biasing of Sequence to Sequence ASR Systems
      (longer introduction)
    • Token-Level Supervised Contrastive Learning for Punctuation Restoration
      (3 minutes introduction)
    • Class-Based Neural Network Language Model For Second-Pass Rescoring In ASR
      (3 minutes introduction)
    • Correcting Automated and Manual Speech Transcription Errors usingWarped Language Models
      (3 minutes introduction)
  • Language Modeling and Text-based Innovations for ASR (3)
    • Text Augmentation for Language Models in High Error Recognition Scenario
      (Oral presentation)
    • On Sampling-Based Training Criteria for Neural Language Modeling
      (Oral presentation)
    • Fast Text-Only Domain Adaptation of RNN-Transducer Prediction Network
      (Oral presentation)
  • Linguistic Components in end-to-end ASR (5)
    • The CSTR System for Multilingual and Code-Switching ASR Challenges for Low Resource Indian Languages
      (Oral presentation)
    • Acoustic Data-Driven Subword Modeling for End-to-End Speech Recognition
      (Oral presentation)
    • Equivalence of Segmental and Neural Transducer Modeling: A Proof of Concept
      (Oral presentation)
    • Modeling Dialectal Variation for Swiss German Automatic Speech Recognition
      (Oral presentation)
    • Out-of-vocabulary Words Detection with Attention and CTC Alignments in an End-to-End ASR System
      (Oral presentation)
  • Low-resource speech recognition (7)
    • Towards unsupervised phone and word segmentation using self-supervised vector-quantized neural networks
      (3 minutes introduction)
    • Speech SimCLR: Combining Contrastive and Reconstruction Objective for Self-supervised Speech Representation Learning
      (3 minutes introduction)
    • Speech SimCLR: Combining Contrastive and Reconstruction Objective for Self-supervised Speech Representation Learning
      (longer introduction)
    • Multilingual transfer of acoustic word embeddings improves when training on languages related to the target zero-resource language
      (3 minutes introduction)
    • Analyzing Speaker Information in Self-Supervised Models to Improve Zero-Resource Speech Processing
      (3 minutes introduction)
    • The Zero Resource Speech Challenge 2021: Spoken language modelling
      (3 minutes introduction)
    • Zero-Shot Federated Learning with New Classes for Audio Classification
      (3 minutes introduction)
  • Miscellanous topics in ASR (3)
    • The Energy and Carbon Footprint of Training End-to-End Speech Recognizers
      (Oral presentation)
    • Radically Old Way of Computing Spectra: Applications in End-to-End ASR
      (Oral presentation)
    • Self-supervised End-to-End ASR for Low Resource L2 Swedish
      (Oral presentation)
  • Multi- and cross-lingual ASR, other topics in ASR (8)
    • Cross-domain Speech Recognition with Unsupervised Character-level Distribution Matching
      (3 minutes introduction)
    • Large-Scale Pre-Training of End-to-End Multi-Talker ASR for Meeting Transcription with Single Distant Microphone
      (3 minutes introduction)
    • Reducing Streaming ASR Model Delay with Self Alignment
      (3 minutes introduction)
    • Reduce and Reconstruct: ASR for Low-Resource Phonetic Languages
      (3 minutes introduction)
    • Reduce and Reconstruct: ASR for Low-Resource Phonetic Languages
      (longer introduction)
    • Listen with Intent: Improving Speech Recognition with Audio-to-Intent Front-End
      (3 minutes introduction)
    • Exploring Targeted Universal Adversarial Perturbations to End-to-end ASR Models
      (3 minutes introduction)
    • Earnings-21: A Practical Benchmark for ASR in the Wild
      (3 minutes introduction)
  • Multi-channel speech enhancement and hearing aids (9)
    • LACOPE: Latency-Constrained Pitch Estimation for Speech Enhancement
      (3 minutes introduction)
    • Multiple Sound Source Localization Based on Interchannel Phase Differences in All Frequencies with Spectral Masks
      (3 minutes introduction)
    • Cancellation of Local Competing Speaker with Near-field Localization for Distributed Ad-Hoc Sensor Network
      (3 minutes introduction)
    • Cancellation of Local Competing Speaker with Near-field Localization for Distributed Ad-Hoc Sensor Network
      (longer introduction)
    • A Deep Learning Method to Multi-Channel Active Noise Control
      (3 minutes introduction)
    • Clarity-2021 challenges: Machine learning challenges for advancing hearing aid processing
      (3 minutes introduction)
    • Clarity-2021 challenges: Machine learning challenges for advancing hearing aid processing
      (longer introduction)
    • Explaining deep learning models for speech enhancement
      (3 minutes introduction)
    • Minimum-Norm Differential Beamforming for Linear Array with Directional Microphones
      (3 minutes introduction)
  • Multimodal systems (10)
    • Direct multimodal few-shot learning of speech and images
      (3 minutes introduction)
    • Talk, Don't Write: A Study of Direct Speech-Based Image Retrieval
      (3 minutes introduction)
    • Cross-Modal Knowledge Distillation Method for Automatic Cued Speech Recognition
      (3 minutes introduction)
    • Cross-Modal Knowledge Distillation Method for Automatic Cued Speech Recognition
      (longer introduction)
    • Attention-Based Keyword Localisation in Speech using Visual Grounding
      (3 minutes introduction)
    • Automatic Lip-reading with Hierarchical Pyramidal Convolution and Self-Attention for Image Sequences with No Word Boundaries
      (3 minutes introduction)
    • Automatic Lip-reading with Hierarchical Pyramidal Convolution and Self-Attention for Image Sequences with No Word Boundaries
      (longer introduction)
    • LiRA: Learning Visual Speech Representations from Audio through Self-supervision
      (3 minutes introduction)
    • End-to-end audio-visual speech recognition for overlapping speech}
      (3 minutes introduction)
    • Audio-Visual Multi-Talker Speech Recognition in A Cocktail Party
      (3 minutes introduction)
  • Neural Network Training Methods and Architectures for ASR (4)
    • Self-paced ensemble learning for speech and audio classification
      (Oral presentation)
    • Multi-Encoder Learning and Stream Fusion for Transformer-Based End-to-End Automatic Speech Recognition
      (Oral presentation)
    • Conditional Independence for Pretext Task Selection in Self-Supervised Speech Representation Learning
      (Oral presentation)
    • Investigating Methods to Improve Language Model Integration for Attention-based Encoder-Decoder ASR Models
      (Oral presentation)
  • Neural network training methods for ASR (9)
    • Towards Lifelong Learning of End-to-end ASR
      (3 minutes introduction)
    • Towards Lifelong Learning of End-to-end ASR
      (longer introduction)
    • Regularizing Word Segmentation by Creating Misspellings
      (3 minutes introduction)
    • Multitask Training with Text Data for End-to-End Speech Recognition
      (3 minutes introduction)
    • Emitting Word Timings with HMM-free End-to-End System in Automatic Speech Recognition
      (3 minutes introduction)
    • Leveraging non-target language resources to improve ASR performance in a target language
      (3 minutes introduction)
    • Leveraging non-target language resources to improve ASR performance in a target language
      (longer introduction)
    • 4-bit Quantization of LSTM-based Speech Recognition Models
      (3 minutes introduction)
    • 4-bit Quantization of LSTM-based Speech Recognition Models
      (longer introduction)
  • Non-Autoregressive Sequential Modeling for Speech Processing (7)
    • Pushing the Limits of Non-Autoregressive Speech Recognition
      (Oral presentation)
    • Layer Pruning on Demand with Intermediate CTC
      (Oral presentation)
    • Real-time End-to-End Monaural Multi-speaker Speech Recognition
      (Oral presentation)
    • TalkNet: Non-Autoregressive Depth-Wise Separable Convolutional Model for Speech Synthesis
      (Oral presentation)
    • WaveGrad 2: Iterative Refinement for Text-to-Speech Synthesis
      (Oral presentation)
    • Align-Denoise: Single-Pass Non-Autoregressive Speech Recognition
      (Oral presentation)
    • VAENAR-TTS: Variational Auto-Encoder based Non-AutoRegressive Text-to-Speech Synthesis
      (Oral presentation)
  • Non-native speech (5)
    • Cross-linguistic Perception of the Japanese Singleton/Geminate Contrast: Korean, Mandarin and Mongolian Compared
      (3 minutes introduction)
    • Detection of Lexical Stress Errors in Non-Native (L2) English with Data Augmentation and Attention
      (longer introduction)
    • Acquisition of prosodic focus marking by three- to six-year-old children learning Mandarin Chinese
      (3 minutes introduction)
    • A neural network-based noise compensation method for pronunciation assessment
      (3 minutes introduction)
    • A Preliminary Study on Discourse Prosody Encoding in L1 and L2 English Spontaneous Narratives
      (3 minutes introduction)
  • Novel neural network architectures for ASR (8)
    • Dynamic Encoder Transducer: A Flexible Solution For Trading Off Accuracy For Latency
      (3 minutes introduction)
    • Librispeech Transducer Model with Internal Language Model Prior Correction
      (3 minutes introduction)
    • A Deliberation-based Joint Acoustic and Text Decoder
      (3 minutes introduction)
    • On the limit of English conversational speech recognition
      (3 minutes introduction)
    • SpeechMoE: Scaling to Large Acoustic Models with Dynamic Routing Mixture of Experts
      (3 minutes introduction)
    • Online Compressive Transformer for End-to-End Speech Recognition
      (3 minutes introduction)
    • End to end transformer-based contextual speech recognition based on pointer network
      (3 minutes introduction)
    • A Comparative Study on Neural Architectures and Training Methods for Japanese Speech Recognition
      (3 minutes introduction)
  • OpenASR20 and Low Resource ASR Development (3)
    • Systems for Low-Resource Speech Recognition Tasks in Open Automatic Speech Recognition and Formosa Speech Recognition Challenges
      (Oral presentation)
    • The TNT Team System Descriptions of Cantonese and Mongolian for IARPA OpenASR20
      (Oral presentation)
    • Combining Hybrid and End-to-end Approaches for the OpenASR20 Challenge
      (Oral presentation)
  • Oriental Language Recognition (3)
    • Dynamic Multi-scale Convolution for Dialect Identification
      (Oral presentation)
    • Language Recognition Based on Unsupervised Pretrained Models
      (Oral presentation)
    • Additive Phoneme-aware Margin Softmax Loss for Language Recognition
      (Oral presentation)
  • Phonation and voicing (4)
    • Synchronic Fortition in Five Romance Languages? A Large Corpus-Based Study of Word-Initial Devoicing
      (Oral presentation)
    • Glottal Stops in Upper Sorbian: a Data-Driven Approach
      (Oral presentation)
    • Glottal Sounds in Korebaju
      (Oral presentation)
    • Automatic classification of phonation types in spontaneous speech: towards a new workflow for the characterization of speakers' voice quality
      (Oral presentation)
  • Phonetics I (1)
    • Prosodic Accommodation in Face-to-face and Telephone Dialogues
      (Oral presentation)
  • Phonetics II (11)
    • Leveraging Real-time MRI for Illuminating Linguistic Velum Action
      (3 minutes introduction)
    • Segmental Alignment of English Syllables with Singleton and Cluster Onsets
      (3 minutes introduction)
    • Exploration of Welsh English Pre-aspiration: How Wide-Spread is it?
      (3 minutes introduction)
    • Revisiting recall effects of filler particles in German and English
      (3 minutes introduction)
    • How reliable are phonetic data collected remotely? Comparison of recording devices and environments on acoustic measurements
      (3 minutes introduction)
    • How reliable are phonetic data collected remotely? Comparison of recording devices and environments on acoustic measurements
      (longer introduction)
    • Quantifying vocal tract shape variation and its acoustic impact: a geometric morphometric approach
      (3 minutes introduction)
    • Speakers coarticulate less when facing real and imagined communicative difficulties: An analysis of read and spontaneous speech from the LUCID corpus
      (3 minutes introduction)
    • Developmental changes of vowel acoustics in adolescents
      (3 minutes introduction)
    • A New Vowel Normalization for Sociophonetics
      (3 minutes introduction)
    • The Pacific Expansion: Optimizing phonetic transcription of archival corpora
      (3 minutes introduction)
  • Privacy-preserving Machine Learning for Audio & Speech Processing (9)
    • Privacy-preserving voice anti-spoofing using secure multi-party computation
      (3 minutes introduction)
    • Configurable Privacy-Preserving Automatic Speech Recognition
      (3 minutes introduction)
    • Adjunct-Emeritus Distillation for Semi-Supervised Language Model Adaptation
      (3 minutes introduction)
    • Communication-Efficient Agnostic Federated Averaging
      (3 minutes introduction)
    • Communication-Efficient Agnostic Federated Averaging
      (longer introduction)
    • Privacy-Preserving Feature Extraction for Cloud-Based Wake Word Verification
      (3 minutes introduction)
    • PATE-AAE: Incorporating Adversarial Autoencoder into Private Aggregation of Teacher Ensembles for Spoken Command Classification
      (3 minutes introduction)
    • PATE-AAE: Incorporating Adversarial Autoencoder into Private Aggregation of Teacher Ensembles for Spoken Command Classification
      (longer introduction)
    • SynthASR: Unlocking Synthetic Data for Speech Recognition
      (3 minutes introduction)
  • Prosodic features and structure (8)
    • An Experiment in Paratone Detection in a Prosodically Annotated EAP Spoken Corpus
      (3 minutes introduction)
    • An Experiment in Paratone Detection in a Prosodically Annotated EAP Spoken Corpus
      (longer introduction)
    • ProsoBeast Prosody Annotation Tool
      (3 minutes introduction)
    • Assessing the Use of Prosody in Constituency Parsing of Imperfect Transcripts
      (3 minutes introduction)
    • In-group advantage in the perception of emotions: Evidence from three varieties of German
      (3 minutes introduction)
    • The LF Model in the Frequency Domain for Glottal Airflow Modelling without Aliasing Distortion
      (3 minutes introduction)
    • Parsing speech for grouping and prominence and the typology of rhythm
      (3 minutes introduction)
    • Leveraging the uniformity framework to examine crosslinguistic similarity for long-lag stops in spontaneous Cantonese-English bilingual speech
      (3 minutes introduction)
  • Resource-constrained ASR (8)
    • Compressing 1D Time-Channel Separable Convolutions using Sparse Random Ternary Matrices
      (3 minutes introduction)
    • Compressing 1D Time-Channel Separable Convolutions using Sparse Random Ternary Matrices
      (longer introduction)
    • Weakly Supervised Construction of ASR Systems from Massive Video Data
      (3 minutes introduction)
    • Weakly Supervised Construction of ASR Systems from Massive Video Data
      (longer introduction)
    • Extremely Low Footprint End-to-End ASR System for Smart Device
      (3 minutes introduction)
    • Tied \& Reduced RNN-T Decoder
      (3 minutes introduction)
    • PQK: Model Compression via Pruning, Quantization, and Knowledge Distillation
      (3 minutes introduction)
    • Collaborative Training of Acoustic Encoders for Speech Recognition
      (3 minutes introduction)
  • Robust and Far-field ASR (3)
    • Self-Attention Channel Combinator Frontend for End-to-End Multichannel Far-field Speech Recognition
      (Oral presentation)
    • ETLT 2021: SHARED TASK ON AUTOMATIC SPEECH RECOGNITION FOR NON-NATIVE CHILDREN’S SPEECH
      (Oral presentation)
    • Learning to Rank Microphones for Distant Speech Recognition
      (Oral presentation)
  • Robust Speaker Recognition (8)
    • Unsupervised Bayesian Adaptation of PLDA for Speaker Verification
      (3 minutes introduction)
    • Variational Information Bottleneck based Regularization for Speaker Recognition
      (3 minutes introduction)
    • SpeakerStew: Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
      (3 minutes introduction)
    • ANTVOICE NEURAL SPEAKER EMBEDDING SYSTEM FOR FFSVC 2020
      (3 minutes introduction)
    • ANTVOICE NEURAL SPEAKER EMBEDDING SYSTEM FOR FFSVC 2020
      (longer introduction)
    • Deep Feature CycleGANs: Speaker Identity Preserving Non-parallel Microphone-Telephone Domain Adaptation for Speaker Verification
      (3 minutes introduction)
    • Joint Feature Enhancement and Speaker Recognition with Multi-Objective Task-Oriented Network
      (3 minutes introduction)
    • Speaker anonymisation using the McAdams coefficient
      (3 minutes introduction)
  • SdSV Challenge 2021: Analysis and Exploration of New Ideas on Short-Duration Speaker Verification (2)
    • Team02 Text-Independent Speaker Verification System for SdSV Challenge 2021
      (longer introduction)
    • The SJTU System for Short-duration Speaker Verification Challenge 2021
      (3 minutes introduction)
  • Search/decoding techniques and confidence measures for ASR (6)
    • LT-LM: a novel non-autoregressive language model for single-shot lattice rescoring
      (3 minutes introduction)
    • Deep neural network calibration for E2E speech recognition system
      (3 minutes introduction)
    • Deep neural network calibration for E2E speech recognition system
      (longer introduction)
    • Residual Energy-Based Models for End-to-End Speech Recognition
      (3 minutes introduction)
    • Insights on Neural Representations for End-to-End Speech Recognition
      (3 minutes introduction)
    • Sequence-level Confidence Classifier for ASR Utterance Accuracy and Application to Acoustic Models
      (3 minutes introduction)
  • Self-supervision and semi-supervision for neural ASR training (5)
    • On the Learning Dynamics of Semi-Supervised Training for ASR
      (3 minutes introduction)
    • Robust wav2vec 2.0: Analyzing Domain Shift in Self-Supervised Pre-Training
      (3 minutes introduction)
    • Semi-Supervision in ASR: Sequential MixMatch and Factorized TTS-Based Augmentation
      (3 minutes introduction)
    • Phonetically Motivated Self-Supervised Speech Representation Learning
      (3 minutes introduction)
    • Improving RNN-T for Domain Scaling Using Semi-Supervised Training with Neural TTS
      (3 minutes introduction)
  • Show and Tell 1 (5)
    • Application for detecting depression, Parkinson’s disease and dysphonic speech
      (3 minutes introduction)
    • Beey: More than a Speech to Text Editor
      (3 minutes introduction)
    • Downsizing of vocal-tract models to line up variations and reduce manufacturing costs
      (3 minutes introduction)
    • ROXANNE Research Platform: Automate criminal investigations and leverage multimodal fusion
      (longer introduction)
    • Advanced semi-blind speaker extraction and tracking implemented in experimental device with revolving dense microphone array
      (3 minutes introduction)
  • Show and Tell 2 (5)
    • Multi-speaker Emotional Text-to-speech Synthesizer
      (3 minutes introduction)
    • Autonomous Robot for Measuring Room Impulse Responses
      (3 minutes introduction)
    • ThemePro 2.0: Showcasing the Role of Thematic Progression in Engaging Human-Computer Interaction
      (3 minutes introduction)
    • ThemePro 2.0: Showcasing the Role of Thematic Progression in Engaging Human-Computer Interaction
      (longer introduction)
    • Audio Segmentation based Conversational Silence Detection for Contact Call Centers
      (3 minutes introduction)
  • Show and Tell 3 (7)
    • MoM: Minutes of Meeting Bot
      (3 minutes introduction)
    • Articulatory Data Recorder: A Framework for Real-Time Articulatory Data Recording
      (3 minutes introduction)
    • The INGENIOUS Multilingual Operations App
      (3 minutes introduction)
    • Digital Einstein Experience: Fast Text-to-Speech for Conversational AI
      (3 minutes introduction)
    • Live Subtitling for BigBlueButton with Open-Source Software
      (3 minutes introduction)
    • Expressive Latvian Speech Synthesis for Dialog Systems
      (3 minutes introduction)
    • Vi STA FAE : A Visual Speech Training Aid with Feedback of Articulatory Efforts
      (3 minutes introduction)
  • Show and Tell 4 (7)
    • Interactive and real-time acoustic measurement tools for speech data acquisition and presentation: Application of an extended member of time stretched pulses
      (3 minutes introduction)
    • Save your Voice: Voice Banking and TTS for Anyone
      (3 minutes introduction)
    • NeMo (Inverse) Text Normalization: From Development To Production
      (3 minutes introduction)
    • NeMo (Inverse) Text Normalization: From Development To Production
      (longer introduction)
    • Automatic Radiology Report Editing through Voice
      (3 minutes introduction)
    • WittyKiddy: Multilingual Spoken Language Learning for Kids
      (3 minutes introduction)
    • Web Interface for estimating articulatory movements in speech production from acoustics and text
      (3 minutes introduction)
  • Single-channel speech enhancement (7)
    • Personalized Speech Enhancement through Self-Supervised Data Augmentation and Purification
      (3 minutes introduction)
    • Speech Denoising with Auditory Models
      (3 minutes introduction)
    • Human Listening and Live Captioning: Multi-Task Training for Speech Enhancement
      (3 minutes introduction)
    • A Maximum Likelihood Approach to SNR-Progressive Learning Using Generalized Gaussian Distribution for LSTM-Based Speech Enhancement
      (3 minutes introduction)
    • WHISPER SPEECH ENHANCEMENT USING JOINT VARIATIONAL AUTOENCODER FOR IMPROVED SPEECH RECOGNITION
      (3 minutes introduction)
    • Speech Denoising without Clean Training Data: a Noise2Noise Approach
      (3 minutes introduction)
    • Speech Enhancement with Topology-enhanced Generative Adversarial Networks (GANs)
      (3 minutes introduction)
  • Source Separation I (2)
    • End-to-End Speech Separation Using Orthogonal Representation in Complex and Real Time-Frequency Domain
      (3 minutes introduction)
    • Dual-Path Filter Network: Speaker-Aware Modeling for Speech Separation
      (3 minutes introduction)
  • Source Separation II (10)
    • Graph-PIT: Generalized permutation invariant training for continuous separation of arbitrary numbers of speakers
      (3 minutes introduction)
    • TEACHER-STUDENT MIXIT FOR UNSUPERVISED AND SEMI-SUPERVISED SPEECH SEPARATION
      (3 minutes introduction)
    • Few shot-learning of new sound classes for target sound extraction
      (3 minutes introduction)
    • AvaTr: One-Shot Speaker Extraction with Transformers
      (3 minutes introduction)
    • Vocal Harmony Separation using Time-domain Neural Networks
      (3 minutes introduction)
    • Vocal Harmony Separation using Time-domain Neural Networks
      (longer introduction)
    • Speaker Verification-Based Evaluation of Single-Channel Speech Separation
      (3 minutes introduction)
    • IMPROVED SPEECH SEPARATION WITH TIME-AND-FREQUENCY CROSS-DOMAIN FEATURE SELECTION
      (3 minutes introduction)
    • Neural Speaker Extraction with Speaker-Speech Cross-Attention Network
      (3 minutes introduction)
    • Deep audio-visual speech separation based on facial motion
      (3 minutes introduction)
  • Source Separation III (3)
    • Combating Reverberation in NTF-based Speech Separation using a Sub-Source Weighted Multichannel Wiener Filter and Linear Prediction
      (Oral presentation)
    • A Hands-on Comparison of DNNs for Dialog Separation Using Transfer Learning from Music Source Separation
      (Oral presentation)
    • GlobalPhone Mix-to-Separate out of 2: A Multilingual 2000 Speakers Mixtures Database for Speech Separation
      (Oral presentation)
  • Source separation, dereverberation and echo cancellation (3)
    • Multi-Stream Gated and Pyramidal Temporal Convolutional Neural Networks for Audio-Visual Speech Separation in Multi-Talker Environments
      (3 minutes introduction)
    • Residual Echo and Noise Cancellation with Feature Attention Module and Multi-domain Loss Function
      (3 minutes introduction)
    • A Deep Learning Approach to Multi-Channel and Multi-Microphone Acoustic Echo Cancellation
      (3 minutes introduction)
  • Speaker Diarization I (3)
    • Three-class Overlapped Speech Detection using a Convolutional Recurrent Neural Network
      (3 minutes introduction)
    • Online Speaker Diarization Equipped with Discriminative Modeling and Guided Inference
      (3 minutes introduction)
    • End-to-end speaker segmentation for overlap-aware resegmentation
      (3 minutes introduction)
  • Speaker Diarization II (9)
    • LEAP Submission for the Third DIHARD Diarization Challenge
      (3 minutes introduction)
    • LEAP Submission for the Third DIHARD Diarization Challenge
      (longer introduction)
    • Investigation of Spatial-Acoustic Features for Overlapping Speech Detection in Multiparty Meetings
      (3 minutes introduction)
    • Target-Speaker Voice Activity Detection with Improved I-Vector Estimation for Unknown Number of Speaker
      (3 minutes introduction)
    • ECAPA-TDNN Embeddings for Speaker Diarization
      (3 minutes introduction)
    • Advances in integration of end-to-end neural and clustering-based diarization for real conversational speech
      (3 minutes introduction)
    • Anonymous speaker clusters: Making distinctions between anonymised speech recordings with clustering interface
      (3 minutes introduction)
    • Anonymous speaker clusters: Making distinctions between anonymised speech recordings with clustering interface
      (longer introduction)
    • Speaker Diarization using Two-pass GMM PLDA Clustering of DNN Embeddings
      (3 minutes introduction)
  • Speaker Recognition: Applications (9)
    • Graph-based Label Propagation for Semi-Supervised Speaker Identification
      (3 minutes introduction)
    • Fusion of Embeddings Networks for Robust Combination of Text Dependent and Independent Speaker Recognition
      (3 minutes introduction)
    • Dr-Vectors: Decision Residual Networks and an Improved Loss for Speaker Recognition
      (3 minutes introduction)
    • Multi-Channel Speaker Verification for Single and Multi-talker Speech
      (3 minutes introduction)
    • Chronological Self-Training for Real-Time Speaker Diarization
      (3 minutes introduction)
    • Presentation matters: Evaluating speaker identification tasks
      (3 minutes introduction)
    • Presentation matters: Evaluating speaker identification tasks
      (longer introduction)
    • Automatic Error Correction for Speaker Embedding Learning with Noisy Labels
      (3 minutes introduction)
    • An Integrated Framework for Two-pass Personalized Voice Trigger
      (3 minutes introduction)
  • Speaker, Language, and Privacy (3)
    • Using Games to Augment Corpora for Language Recognition and Confusability
      (Oral presentation)
    • Fair Voice Biometrics: Impact of Demographic Imbalance on Group Fairness in Speaker Recognition
      (Oral presentation)
    • Knowledge Distillation from Multi-Modality to Single-Modality for Person Verification
      (Oral presentation)
  • Speech and audio analysis (4)
    • Extending the Fullband E-Model Towards Background Noise, Bursty Packet Loss, and Conversational Degradations
      (Oral presentation)
    • ORCA-SLANG: An Automatic Multi-Stage Semi-Supervised Deep Learning Framework for Large-Scale Killer Whale Call Type Identification
      (Oral presentation)
    • Non-Intrusive Speech Quality Assessment with Transfer Learning and Subject-specific Scaling
      (Oral presentation)
    • Audio Retrieval with Natural Language Queries
      (Oral presentation)
  • Speech coding and privacy (9)
    • NU-Wave: A Diffusion Probabilistic Model for Neural Audio Upsampling
      (3 minutes introduction)
    • NU-Wave: A Diffusion Probabilistic Model for Neural Audio Upsampling
      (longer introduction)
    • WSRGlow: A Glow-based Waveform Generative Model for Audio Super-Resolution
      (3 minutes introduction)
    • Multi-channel Opus compression for far-field automatic speech recognition with a fixed bitrate budget
      (longer introduction)
    • Effects of Prosodic Variations on Accidental Triggers of a commercial Voice Assistant
      (3 minutes introduction)
    • Improving the expressiveness of neural vocoding with non-affine Normalizing Flows
      (3 minutes introduction)
    • A Two-stage Approach to Speech Bandwidth Extension
      (3 minutes introduction)
    • Development of a Psychoacoustic Loss Function for the Deep Neural Network (DNN)-Based Speech Coder
      (3 minutes introduction)
    • Protecting gender and identity with disentangled speech representations
      (3 minutes introduction)
  • Speech enhancement and coding (2)
    • End-to-end Optimized Multi-stage Vector Quantization of Spectral Envelopes for Speech and Audio Coding
      (Oral presentation)
    • Fusion-Net: Time-Frequency Information Fusion Y-Network for Speech Enhancement
      (Oral presentation)
  • Speech enhancement and intelligibility (12)
    • Funnel Deep Complex U-net for Phase-Aware Speech Enhancement
      (3 minutes introduction)
    • Perceptual Contributions of Vowels and Consonant-vowel Transitions in Understanding Time-compressed Mandarin Sentences
      (3 minutes introduction)
    • Speech Enhancement with Weakly Labelled Data from AudioSet
      (3 minutes introduction)
    • Improving Perceptual Quality by Phone-Fortified Perceptual Loss using Wasserstein Distance for Speech Enhancement
      (3 minutes introduction)
    • MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement
      (3 minutes introduction)
    • A Spectro-Temporal Glimpsing Index (STGI) for Speech Intelligibility Prediction
      (3 minutes introduction)
    • A Spectro-Temporal Glimpsing Index (STGI) for Speech Intelligibility Prediction
      (longer introduction)
    • Self-Supervised Learning Based Phone-Fortified Speech Enhancement
      (3 minutes introduction)
    • Incorporating Embedding Vectors from a Human Mean-Opinion Score Prediction Model for Monaural Speech Enhancement
      (3 minutes introduction)
    • Incorporating Embedding Vectors from a Human Mean-Opinion Score Prediction Model for Monaural Speech Enhancement
      (longer introduction)
    • Restoring degraded speech via a modified diffusion model
      (3 minutes introduction)
    • Restoring degraded speech via a modified diffusion model
      (longer introduction)
  • Speech Localization, Enhancement, and Quality Assessment (4)
    • PILOT: Introducing Transformers for Probabilistic Sound Event Localization
      (3 minutes introduction)
    • Assessment of von Mises--Bernoulli Deep Neural Network in Sound Source Localization
      (3 minutes introduction)
    • Far-field Speaker Localization and Adaptive GLMB Tracking
      (3 minutes introduction)
    • Far-field Speaker Localization and Adaptive GLMB Tracking
      (longer introduction)
  • Speech perception I (2)
    • Prosodic disambiguation using chironomic stylization of intonation for native and non-native speakers
      (Oral presentation)
    • Extracting Different Levels of Speech Information from EEG Using an LSTM-Based Model
      (Oral presentation)
  • Speech perception II (9)
    • Exploring the Potential of Lexical Paraphrases for Mitigating Noise-Induced Comprehension Errors
      (3 minutes introduction)
    • Exploring the Potential of Lexical Paraphrases for Mitigating Noise-Induced Comprehension Errors
      (longer introduction)
    • VocalTurk: Exploring Feasibility of Crowdsourced Speaker Identification
      (3 minutes introduction)
    • Relationships between Perceptual Distinctiveness, Articulatory Complexity and Functional Load in Speech Communication
      (3 minutes introduction)
    • Human spoofing detection performance on degraded speech
      (3 minutes introduction)
    • Human spoofing detection performance on degraded speech
      (longer introduction)
    • Reliable estimates of interpretable cue effects with Active Learning in psycholinguistic research
      (3 minutes introduction)
    • Towards the explainability of Multimodal Speech Emotion Recognition
      (3 minutes introduction)
    • Investigating the Impact of Spectral and Temporal Degradation on End-to-End Automatic Speech Recognition Performance
      (3 minutes introduction)
  • Speech production I (4)
    • Towards the prediction of the vocal tract shape from the sequence of phonemes to be articulated
      (Oral presentation)
    • Comparison of the finite element method, the multimodal method and thetransmission-line model for the computation of vocal tract transfer functions
      (Oral presentation)
    • Importance of Parasagittal Sensor Information in Tongue Motion Capture through a Diphonic Analysis
      (Oral presentation)
    • Changes in glottal source parameter values with light to moderate physical load
      (Oral presentation)
  • Speech production II (6)
    • A Simplified Model for the Vocal Tract of [s] with Inclined Incisors
      (3 minutes introduction)
    • Vocal-tract models to visualize the airstream of human breath and droplets while producing speech
      (3 minutes introduction)
    • Using Transposed Convolution for Articulatory-to-Acoustic Conversion from Real-Time MRI Data
      (3 minutes introduction)
    • Inhalations in speech: acoustic and physiological characteristics
      (3 minutes introduction)
    • Take a breath: Respiratory sounds improve recollection in synthetic speech
      (3 minutes introduction)
    • Mixture of orthogonal sequences made from extended time-stretched pulses enables measurement of involuntary voice fundamental frequency response to pitch perturbation
      (3 minutes introduction)
  • Speech Recognition of Atypical Speech (11)
    • Automatic Speech Recognition of Disordered Speech: Personalized models outperforming human listeners on short phrases
      (Oral presentation)
    • Investigating the Utility of Multimodal Conversational Technology and Audiovisual Analytic Measures for the Assessment and Monitoring of Amyotrophic Lateral Sclerosis at Scale
      (Oral presentation)
    • Handling acoustic variation in dysarthric speech recognition systems through model combination
      (Oral presentation)
    • Adversarial Data Augmentation for Disordered Speech Recognition
      (Oral presentation)
    • Variational Auto-Encoder Based Variability Encoding for Dysarthric Speech Recognition
      (Oral presentation)
    • Bayesian Parametric and Architectural Domain Adaptation of LF-MMI Trained TDNNs for Elderly and Dysarthric Speech Recognition
      (Oral presentation)
    • A Voice-Activated Switch for Persons with Motor and Speech Impairments: Isolated-Vowel Spotting Using Neural Networks
      (Oral presentation)
    • Conformer Parrotron: a Faster and Stronger End-to-end SpeechConversion and Recognition Model for Atypical Speech
      (Oral presentation)
    • Disordered Speech Data Collection: Lessons Learned at 1 Million Utterances from Project Euphonia
      (Oral presentation)
    • Comparing Supervised Models And Learned Speech Representations For Classifying Intelligibility Of Disordered Speech On Selected Phrases
      (Oral presentation)
    • Analysis and Tuning of a Voice Assistant System for Dysfluent Speech
      (Oral presentation)
  • Speech signal analysis and representation I (12)
    • Estimating articulatory movements in speech production with transformer networks
      (3 minutes introduction)
    • Estimating articulatory movements in speech production with transformer networks
      (longer introduction)
    • Speech Decomposition based on a Hybrid Speech Model and Optimal Segmentation
      (3 minutes introduction)
    • Dropout Regularization for Self-Supervised Learning of Transformer Encoder Speech Representation
      (3 minutes introduction)
    • Noise robust pitch stylization using minimum mean absolute error criterion
      (3 minutes introduction)
    • An Attribute-Aligned Strategy for Learning Speech Representation
      (3 minutes introduction)
    • Raw Speech-to-Articulatory Inversion by Temporal Filtering and Decimation
      (3 minutes introduction)
    • Unsupervised Training of a DNN-based Formant Tracker
      (3 minutes introduction)
    • Unsupervised Training of a DNN-based Formant Tracker
      (longer introduction)
    • Synchronising speech segments with musical beats in Mandarin and English singing
      (3 minutes introduction)
    • Pitch contour separation from overlapping speech
      (3 minutes introduction)
    • Pitch contour separation from overlapping speech
      (longer introduction)
  • Speech signal analysis and representation II (4)
    • A Benchmark of Dynamical Variational Autoencoders applied to Speech Spectrogram Modeling
      (Oral presentation)
    • Fricative Phoneme Detection Using Deep Neural Networks and its Comparison to Traditional Methods
      (Oral presentation)
    • Identification of F1 and F2 in speech using modified zero frequency filtering
      (Oral presentation)
    • Phoneme-to-audio alignment with recurrent neural networks for speaking and singing voice
      (Oral presentation)
  • Speech Synthesis: Linguistic processing, paradigms and other topics (8)
    • UNSUPERVISED LEARNING OF DISENTANGLED SPEECH CONTENT AND STYLE REPRESENTATION
      (3 minutes introduction)
    • Label Embedding for Chinese Grapheme-to-Phoneme Conversion
      (3 minutes introduction)
    • Improving Polyphone Disambiguation for Mandarin Chinese by Combining Mix-pooling Strategy and Window-based Attention
      (3 minutes introduction)
    • Improving Polyphone Disambiguation for Mandarin Chinese by Combining Mix-pooling Strategy and Window-based Attention
      (longer introduction)
    • Polyphone Disambiguition in Mandarin Chinese with Semi-Supervised Learning
      (3 minutes introduction)
    • A Neural-Network-Based Approach to Identifying Speakers in Novels
      (3 minutes introduction)
    • Dynamically Adaptive Machine Speech Chain Inference for TTS in Noisy Environment: Listen and Speak Louder
      (3 minutes introduction)
    • Dynamically Adaptive Machine Speech Chain Inference for TTS in Noisy Environment: Listen and Speak Louder
      (longer introduction)
  • Speech Synthesis: Neural Waveform Generation (6)
    • Glow-WaveGAN: Learning Speech Representations from GAN-based Variational Auto-Encoder For High Fidelity Flow-based Speech Synthesis
      (3 minutes introduction)
    • Harmonic WaveGAN: GAN-Based Speech Waveform Generation Model with Harmonic Structure Discriminator
      (3 minutes introduction)
    • Fre-GAN: Adversarial Frequency-consistent Audio Synthesis
      (3 minutes introduction)
    • GANSpeech: Adversarial Training for High-Fidelity Multi-Speaker Speech Synthesis
      (3 minutes introduction)
    • Continuous Wavelet Vocoder-based Decomposition of Parametric Speech Waveform Synthesis
      (3 minutes introduction)
    • High-fidelity Parallel WaveGAN with Multi-band Harmonic-plus-Noise Model
      (3 minutes introduction)
  • Speech Synthesis: Other topics I (4)
    • Conversion of airborne to bone-conducted speech with deep neural networks
      (Oral presentation)
    • T5G2P: Using Text-to-Text Transfer Transformer for Grapheme-to-Phoneme Conversion
      (Oral presentation)
    • Evaluating the Extrapolation Capabilities of Neural Vocoders to Extreme Pitch Values
      (Oral presentation)
    • A Systematic Review and Analysis of Multilingual Data Strategies in Text-to-Speech for Low-Resource Languages
      (Oral presentation)
  • Speech Synthesis: Prosody Modeling I (6)
    • Improving multi-speaker TTS prosody variance with a residual encoder and normalizing flows
      (3 minutes introduction)
    • Phoneme Duration Modeling Using Speech Rhythm-Based Speaker Embeddings for Multi-Speaker Speech Synthesis
      (3 minutes introduction)
    • Fine-grained Prosody Modeling in Neural Speech Synthesis using ToBI Representation
      (3 minutes introduction)
    • Intra-Sentential Speaking Rate Control in Neural Text-To-Speech for Automatic Dubbing
      (3 minutes introduction)
    • Intra-Sentential Speaking Rate Control in Neural Text-To-Speech for Automatic Dubbing
      (longer introduction)
    • Applying the Information Bottleneck Principle to Prosodic Representation Learning
      (3 minutes introduction)
  • Speech Synthesis: Prosody Modeling II (3)
    • Alternate Endings: Improving Prosody for Incremental Neural TTS with Predicted Future Text Input
      (Oral presentation)
    • Exploring emotional prototypes in a high dimensional TTS latent space
      (Oral presentation)
    • ADEPT: A Dataset for Evaluating Prosody Transfer
      (Oral presentation)
  • Speech Synthesis: Singing, Multimodal, Crosslingual Synthesis (8)
    • Cross-lingual Low Resource Speaker Adaptation Using Phonological Features
      (3 minutes introduction)
    • Improve Cross-Lingual Text-To-Speech Synthesis on Monolingual Corpora with Pitch Contour Information
      (3 minutes introduction)
    • EfficientSing: A Chinese Singing Voice Synthesis System Using Duration-Free Acoustic Model and HiFi-GAN Vocoder
      (3 minutes introduction)
    • Cross-lingual Speaker Adaptation using Domain Adaptation and Speaker Consistency Loss for Text-To-Speech Synthesis
      (3 minutes introduction)
    • Cross-lingual Speaker Adaptation using Domain Adaptation and Speaker Consistency Loss for Text-To-Speech Synthesis
      (longer introduction)
    • Investigating Contributions of Speech and Facial Landmarks for Talking Head Generation
      (3 minutes introduction)
    • Speech2Video: Cross-Modal Distillation for Speech to Video Generation
      (3 minutes introduction)
    • Speech2Video: Cross-Modal Distillation for Speech to Video Generation
      (longer introduction)
  • Speech Synthesis: Speaking Style and Emotion (7)
    • STYLER: Style Factor Modeling with Rapidity and Robustness via Speech Decomposition for Expressive and Controllable Neural Text to Speech
      (3 minutes introduction)
    • Controllable Context-Aware Conversational Speech Synthesis
      (3 minutes introduction)
    • Expressive Text-to-Speech using Style Tag
      (3 minutes introduction)
    • SponSpeech: Adaptive Text to Speech for Spontaneous Style
      (3 minutes introduction)
    • Towards Multi-Scale Style Control for Expressive Speech Synthesis
      (3 minutes introduction)
    • Cross-speaker Style Transfer with Prosody Bottleneck in Neural Speech Synthesis
      (3 minutes introduction)
    • Synthesis of expressive speaking styles with limited training data in a multi-speaker, prosody-controllable sequence-to-sequence architecture
      (3 minutes introduction)
  • Speech Synthesis: tools, data, evaluation (8)
    • Spectral and Latent Speech Representation Distortion for TTS Evaluation
      (3 minutes introduction)
    • RyanSpeech: A Corpus for Conversational Text-to-Speech Synthesis
      (3 minutes introduction)
    • Comparing Speech Enhancement Techniques for Voice Adaptation-Based Speech Synthesis
      (3 minutes introduction)
    • Perception of social speaker characteristics in synthetic speech
      (3 minutes introduction)
    • Hi-Fi Multi-Speaker English TTS Dataset
      (3 minutes introduction)
    • Utilizing Self-supervised Representations for MOS Prediction
      (3 minutes introduction)
    • Utilizing Self-supervised Representations for MOS Prediction
      (longer introduction)
    • KazakhTTS: An Open-Source Kazakh Text-to-Speech Synthesis Dataset
      (3 minutes introduction)
  • Speech Synthesis: Toward End-to-End Synthesis I (7)
    • Federated Learning with Dynamic Transformer on Text to Speech
      (3 minutes introduction)
    • Zero-Shot Text-to-Speech for Text-Based Insertion in Audio Narration
      (3 minutes introduction)
    • Diff-TTS: A Denoising Diffusion Model for Text-to-Speech
      (3 minutes introduction)
    • A Universal Multi-Speaker Multi-Style Text-to-Speech via Disentangled Representation Learning based on Rényi Divergence Minimization
      (3 minutes introduction)
    • Reinforce-Aligner: Reinforcement Alignment Search for Robust End-to-End Text-to-Speech
      (3 minutes introduction)
    • Triple M: A Practical Text-to-speech Synthesis System With Multi-guidance Attention And Multi-band Multi-time LPCNet
      (3 minutes introduction)
    • SC-GlowTTS: an Efficient Zero-Shot Multi-Speaker Text-To-Speech Model
      (3 minutes introduction)
  • Speech Synthesis: Toward End-to-End Synthesis II (8)
    • TacoLPCNet: Fast and Stable TTS by Conditioning LPCNet on Mel Spectrogram Predictions
      (3 minutes introduction)
    • Phonetic and Prosodic Information Estimation Using Neural Machine Translation for Genuine Japanese End-to-End Text-to-Speech
      (3 minutes introduction)
    • Phonetic and Prosodic Information Estimation Using Neural Machine Translation for Genuine Japanese End-to-End Text-to-Speech
      (longer introduction)
    • Information Sieve: Content Leakage Reduction in End-to-End Prosody Transfer for Expressive Speech Synthesis
      (3 minutes introduction)
    • Information Sieve: Content Leakage Reduction in End-to-End Prosody Transfer for Expressive Speech Synthesis
      (longer introduction)
    • PnG BERT: Augmented BERT on Phonemes and Graphemes for Neural TTS
      (3 minutes introduction)
    • Speed up training with variable length inputs by efficient batching strategies
      (3 minutes introduction)
    • Speed up training with variable length inputs by efficient batching strategies
      (longer introduction)
  • Speech type classification and diagnosis (8)
    • A Multi-Branch Deep Learning Network for Automated Detection of COVID-19
      (3 minutes introduction)
    • RW-Resnet: A Novel Speech Anti-Spoofing Model Using Raw Waveform
      (3 minutes introduction)
    • Fake Audio Detection in Resource-constrained Settings using Microfeatures
      (3 minutes introduction)
    • Coughing-based Recognition of Covid-19 with Spatial Attentive ConvLSTM Recurrent Neural Networks
      (3 minutes introduction)
    • Coughing-based Recognition of Covid-19 with Spatial Attentive ConvLSTM Recurrent Neural Networks
      (longer introduction)
    • Knowledge Distillation for Singing Voice Detection
      (3 minutes introduction)
    • Deep Spectral-Cepstral Fusion for Shouted and Normal Speech Classification
      (3 minutes introduction)
    • Automatic Detection of Shouted Speech Segments in Indian News Debates
      (3 minutes introduction)
  • Spoken Dialogue Systems I (2)
    • Act-Aware Slot-Value Predicting in Multi-Domain Dialogue State Tracking
      (3 minutes introduction)
    • Domain-Specific Multi-Agent Dialog Policy Learning in Multi-Domain Task-Oriented Scenarios
      (3 minutes introduction)
  • Spoken Dialogue Systems II (5)
    • Contextualized Attention-based Knowledge Transfer for Spoken Conversational Question Answering
      (3 minutes introduction)
    • Injecting Descriptive Meta-information into Pre-trained Language Models with Hypernetworks
      (3 minutes introduction)
    • Causal Confusion Reduction for Robust Multi-Domain Dialogue Policy
      (3 minutes introduction)
    • Timing Generating Networks: Neural Network based Precise Turn-taking Timing Prediction in Multiparty Conversation
      (3 minutes introduction)
    • Phoneme-BERT: Joint Language Modelling of Phoneme Sequence and ASR Transcript
      (3 minutes introduction)
  • Spoken Language Processing I (7)
    • SmallER: Scaling Neural Entity Resolution for Edge Devices
      (3 minutes introduction)
    • Disfluency Detection with Unlabeled Data and Small BERT Models
      (3 minutes introduction)
    • Discriminative Self-training for Punctuation Prediction
      (3 minutes introduction)
    • A noise robust method for word-level pronunciation assessment
      (3 minutes introduction)
    • Multimodal Speech Summarization through Semantic Concept Learning
      (3 minutes introduction)
    • Enhancing Semantic Understanding with Self-supervised Methods for Abstractive Dialogue Summarization
      (3 minutes introduction)
    • Enhancing Semantic Understanding with Self-supervised Methods for Abstractive Dialogue Summarization
      (longer introduction)
  • Spoken Language Processing II (2)
    • Impact of Encoding and Segmentation Strategies on End-to-End Simultaneous Speech Translation
      (Oral presentation)
    • Speaker transition patterns in three-party conversation: evidence from English, Estonian and Swedish
      (Oral presentation)
  • Spoken Language Understanding I (8)
    • Sequential End-to-End Intent and Slot Label Classification and Localization
      (3 minutes introduction)
    • DEXTER: Deep Encoding of External Knowledge for Named Entity Recognition in Virtual Assistants
      (3 minutes introduction)
    • DEXTER: Deep Encoding of External Knowledge for Named Entity Recognition in Virtual Assistants
      (longer introduction)
    • A Context-Aware Hierarchical BERT Fusion Network for Multi-turn Dialog Act Detection
      (3 minutes introduction)
    • Pre-training for Spoken Language Understanding with Joint Textual and Phonetic Representation Learning
      (3 minutes introduction)
    • Integrating Dialog History into End-to-End Spoken Language Understanding Systems
      (3 minutes introduction)
    • Coreference Augmentation for Multi-Domain Task-Oriented Dialogue State Tracking
      (3 minutes introduction)
    • Rethinking End-to-End Evaluation of Decomposable Tasks: A Case Study on Spoken Language Understanding
      (3 minutes introduction)
  • Spoken Language Understanding II (3)
    • Augmenting Slot Values and Contexts for Spoken Language Understanding with Pretrained Models
      (3 minutes introduction)
    • END-to-END Cross-Lingual Spoken Language Understanding Model with Multilingual Pretraining
      (3 minutes introduction)
    • Factorization-Aware Training of Transformers for Natural Language Understanding On the Edge
      (3 minutes introduction)
  • Spoken machine translation (12)
    • Subtitle Translation as Markup Translation
      (3 minutes introduction)
    • Large-Scale Self- and Semi-Supervised Learning for Speech Translation
      (3 minutes introduction)
    • CoVoST 2 and Massively Multilingual Speech Translation
      (3 minutes introduction)
    • AlloST: Low-resource Speech Translation without Source Transcription
      (3 minutes introduction)
    • Weakly-supervised Speech-to-text Mapping with Visually Connected Non-parallel Speech-text Data using Cyclic Partially-aligned Transformer
      (3 minutes introduction)
    • Weakly-supervised Speech-to-text Mapping with Visually Connected Non-parallel Speech-text Data using Cyclic Partially-aligned Transformer
      (longer introduction)
    • End-to-end Speech Translation via Cross-modal Progressive Training
      (3 minutes introduction)
    • End-to-end Speech Translation via Cross-modal Progressive Training
      (longer introduction)
    • Towards simultaneous machine interpretation
      (3 minutes introduction)
    • Lexical Modeling of ASR Errors for Robust Speech Translation
      (3 minutes introduction)
    • Lexical Modeling of ASR Errors for Robust Speech Translation
      (longer introduction)
    • Effects of Feature Scaling and Fusion on Sign Language Translation
      (3 minutes introduction)
  • Spoken Term Detection & Voice Search (9)
    • Do Acoustic Word Embeddings Capture Phonological Similarity? An Empirical Study
      (3 minutes introduction)
    • Paraphrase Label Alignment for Voice Application Retrieval in Spoken Language Understanding
      (3 minutes introduction)
    • Paraphrase Label Alignment for Voice Application Retrieval in Spoken Language Understanding
      (longer introduction)
    • Streaming Transformer for Hardware Efficient Voice Trigger Detection and False Trigger Mitigation
      (3 minutes introduction)
    • Few-Shot Keyword Spotting in Any Language
      (3 minutes introduction)
    • Text Anchor Based Metric Learning for Small-footprint Keyword Spotting
      (3 minutes introduction)
    • A meta-learning approach for user-defined spoken term classification with varying classes and examples
      (3 minutes introduction)
    • Auxiliary Sequence Labeling Tasks for Disfluency Detection
      (3 minutes introduction)
    • Keyword Transformer: A Self-Attention Model for Keyword Spotting
      (3 minutes introduction)
  • Streaming for ASR/RNN Transducers (7)
    • Super-Human Performance in Online Low-latency Recognition of Conversational Speech
      (3 minutes introduction)
    • Multiple Softmax Architecture for Streaming Multilingual End-to-End ASR Systems
      (3 minutes introduction)
    • An Efficient Streaming Non-Recurrent On-Device End-to-End Model with Improvements to Rare-Word Modeling
      (3 minutes introduction)
    • Reducing Exposure Bias in Training Recurrent Neural Network Transducers
      (3 minutes introduction)
    • Bridging the gap between streaming and non-streaming ASR systems by distilling ensembles of CTC and RNN-T models
      (3 minutes introduction)
    • Bridging the gap between streaming and non-streaming ASR systems by distilling ensembles of CTC and RNN-T models
      (longer introduction)
    • Mixture Model Attention: Flexible Streaming and Non-Streaming Automatic Speech Recognition
      (3 minutes introduction)
  • Target speaker detection, localization and separation (5)
    • Auxiliary loss function for target speech extraction and recognition with weak supervision based on speaker characteristics
      (Oral presentation)
    • Universal Speaker Extraction in the Presence and Absence of Target Speakers for Speech of One and Two Talkers
      (Oral presentation)
    • Using X-vectors for Speech Activity Detection in Broadcast Streams
      (Oral presentation)
    • Time Delay Estimation for Speaker Localization Using CNN-Based Parametrized GCC-PHAT Features
      (Oral presentation)
    • Real-time Speaker counting in a cocktail party scenario using Attention-guided Convolutional Neural Network
      (Oral presentation)
  • The ADReSSo Challenge: Detecting cognitive decline using speech only (7)
    • Influence of the Interviewer on the Automatic Assessment of Alzheimer's Disease in the Context of the ADReSSo Challenge
      (3 minutes introduction)
    • WavBERT: Exploiting Semantic and Non-semantic Speech using Wav2vec and BERT for Dementia Detection
      (3 minutes introduction)
    • Alzheimer Disease Recognition Using Speech-Based Embeddings From Pre-Trained Models
      (3 minutes introduction)
    • Exploring using the outputs of different automatic speech recognition paradigms for acoustic- and BERT-based Alzheimer's Dementia detection through spontaneous speech
      (3 minutes introduction)
    • Automatic Detection of Alzheimer's Disease Using Spontaneous Speech Only
      (3 minutes introduction)
    • Modular Multi-Modal Attention Network for Alzheimer's Disease Detection Using Patient Audio and Language Data
      (3 minutes introduction)
    • Modular Multi-Modal Attention Network for Alzheimer's Disease Detection Using Patient Audio and Language Data
      (longer introduction)
  • The First DiCOVA Challenge: Diagnosis of COVid-19 using Acoustics (6)
    • Diagnosis of COVID-19 using Auditory Acoustic Cues
      (Oral presentation)
    • Classification of COVID-19 from Cough Using Autoregressive Predictive Coding Pretraining and Spectral Data Augmentation
      (Oral presentation)
    • The DiCOVA 2021 Challenge -- An Encoder-Decoder Approach for COVID-19 Recognition from Coughing Audio
      (Oral presentation)
    • COVID-19 Detection from Spectral features on the DiCOVA Dataset
      (Oral presentation)
    • Cough-based COVID-19 Detection with Contextual Attention Convolutional Neural Networks and Gender Information
      (Oral presentation)
    • Contrastive Learning of Cough Descriptors for Automatic COVID-19 Preliminary Diagnosis
      (Oral presentation)
  • The INTERSPEECH 2021 Computational Paralinguistics Challenge (ComParE) - COVID-19 Cough, COVID-19 Speech, Escalation & Primates (8)
    • Transfer Learning-Based Cough Representations for Automatic Detection of COVID-19
      (Oral presentation)
    • Transfer Learning and Data Augmentation Techniques to the COVID-19 Identification Tasks in ComParE 2021
      (Oral presentation)
    • Vision Transformer for Audio-based Primates Classification and COVID Detection
      (Oral presentation)
    • Deep-learning-based central African primate species classification with MixUp and SpecAugment
      (Oral presentation)
    • Introducing a Central African Primate Vocalisation Dataset for Automated Species Classification
      (Oral presentation)
    • Multi-Attentive Detection of the Spider Monkey Whinny in the (Actual) Wild
      (Oral presentation)
    • Identifying Conflict Escalation and Primates by Using Ensemble X-vectors and Fisher Vector Features
      (Oral presentation)
    • Ensemble-within-ensemble classification for escalation prediction from speech
      (Oral presentation)
  • Tools, corpora and resources (11)
    • The Multilingual TEDx Corpus for Speech Recognition and Translation
      (3 minutes introduction)
    • GigaSpeech: An Evolving, Multi-domain ASR Corpus with 10,000 Hours of Transcribed Audio
      (3 minutes introduction)
    • AusKidTalk: An Auditory-Visual Corpus of 3- to 12-year-old Australian Children’s Speech
      (3 minutes introduction)
    • Human-in-the-Loop Efficiency Analysis for Binary Classification in Edyson
      (3 minutes introduction)
    • Annotation Confidence vs. Training Sample Size: Trade-off Solution for Partially-Continuous Categorical Emotion Recognition
      (3 minutes introduction)
    • Europarl-ASR: A Large Corpus of Parliamentary Debates for Streaming ASR Benchmarking and Speech Data Filtering/Verbatimization
      (3 minutes introduction)
    • Towards Automatic Speech to Sign Language Generation
      (3 minutes introduction)
    • Towards Automatic Speech to Sign Language Generation
      (longer introduction)
    • kosp2e: Korean Speech to English Translation Corpus
      (3 minutes introduction)
    • speechocean762: An Open-Source Non-native English Speech Corpus For Pronunciation Assessment
      (3 minutes introduction)
    • speechocean762: An Open-Source Non-native English Speech Corpus For Pronunciation Assessment
      (longer introduction)
  • Topics in ASR: Adaptation, transfer learning, children's speech, and low-resource settings (9)
    • Semantic Data Augmentation for End-to-End Mandarin Speech Recognition
      (3 minutes introduction)
    • Low Resource German ASR with Untranscribed Data Spoken by Non-native Children - INTERSPEECH 2021 Shared Task SPAPL System
      (3 minutes introduction)
    • Low Resource German ASR with Untranscribed Data Spoken by Non-native Children - INTERSPEECH 2021 Shared Task SPAPL System
      (longer introduction)
    • Speaker normalization using Joint Variational Autoencoder
      (3 minutes introduction)
    • The TAL system for the INTERSPEECH2021 Shared Task on Automatic Speech Recognition for Non-Native Children’s Speech
      (3 minutes introduction)
    • The TAL system for the INTERSPEECH2021 Shared Task on Automatic Speech Recognition for Non-Native Children’s Speech
      (longer introduction)
    • Zero-shot Cross-Lingual Phonetic Recognition with External Language Embedding
      (3 minutes introduction)
    • Best of Both Worlds: Robust Accented Speech Recognition with Adversarial Transfer Learning
      (3 minutes introduction)
    • Extending Pronunciation Dictionary with Automatically Detected Word Mispronunciations to Improve PAII's System for Interspeech 2021 Non-Native Child English Close Track ASR Challenge
      (3 minutes introduction)
  • Topics in ASR: Robustness, feature extraction, and far-field ASR (8)
    • End-to-End Spelling Correction Conditioned on Acoustic Feature for Code-switching Speech Recognition
      (3 minutes introduction)
    • Phoneme Recognition through Fine Tuning of Phonetic Representations: a Case Study on Luhya Language Varieties
      (3 minutes introduction)
    • Speech Acoustic Modelling using Raw Source and Filter Components
      (3 minutes introduction)
    • IR-GAN: Room impulse response generator for far-field speech recognition
      (3 minutes introduction)
    • Multi-Channel Transformer Transducer for Speech Recognition
      (3 minutes introduction)
    • Multi-Channel Transformer Transducer for Speech Recognition
      (longer introduction)
    • Leveraging Phone Mask Training for Phonetic-Reduction-Robust E2E Uyghur Speech Recognition
      (3 minutes introduction)
    • Raw Waveform Encoder with Multi-Scale Globally Attentive Locally Recurrent Networks for End-to-End Speech Recognition
      (3 minutes introduction)
  • Tutorials (8)
    • Intonation Transcription and Modelling in Research and Speech Technology Applications
    • Neural target speech extraction
    • Speech Recognition with Next-Generation Kaldi (K2, Lhotse, Icefall)
    • An Introduction to Automatic Differentiation with Weighted Finite-State Automata
    • SpeechBrain: Unifying Speech Technologies and Deep Learning With an Open Source Toolkit
    • SpeechBrain: Unifying Speech Technologies and Deep Learning With an Open Source Toolkit
    • SpeechBrain: Unifying Speech Technologies and Deep Learning With an Open Source Toolkit
    • Concept to Code: Semi-Supervised End-To-End Approaches For Speech Recognition
  • Voice activity detection (5)
    • Unsupervised Representation Learning for Speech Activity Detection in the Fearless Steps Challenge 2021
      (Oral presentation)
    • The Application of Learnable STRF Kernels to the 2021 Fearless Steps Phase-03 SAD Challenge
      (Oral presentation)
    • Speech Activity Detection Based on Multilingual Speech Recognition System
      (Oral presentation)
    • Voice Activity Detection With Teacher-Student Domain Emulation
      (Oral presentation)
    • EML Online Speech Activity Detection for Fearless Steps Challenge Phase-III
      (Oral presentation)
  • Voice activity detection and keyword spotting (10)
    • Attention-based cross-modal fusion for audio-visual voice activity detection in musical video streams
      (3 minutes introduction)
    • Noise-Tolerant Self-Supervised Learning for Audio-Visual Voice Activity Detection
      (3 minutes introduction)
    • Multi-Channel VAD for Transcription of Group Discussion
      (3 minutes introduction)
    • Multi-Channel VAD for Transcription of Group Discussion
      (longer introduction)
    • Audio-Visual Information Fusion Using Cross-modal Teacher-Student Learning for Voice Activity Detection in Realistic Environments
      (3 minutes introduction)
    • Enrollment-less training for personalized voice activity detection
      (3 minutes introduction)
    • FastICARL: Fast Incremental Classifier and Representation Learning with Efficient Budget Allocation in Audio Sensing Applications
      (3 minutes introduction)
    • End-to-End Transformer-Based Open-Vocabulary Keyword Spotting with Location-Guided Local Attention
      (3 minutes introduction)
    • Segmental Contrastive Predictive Coding for Unsupervised Word Segmentation
      (3 minutes introduction)
    • A Lightweight Framework for Online Voice Activity Detection in the Wild
      (3 minutes introduction)
  • Voice and voicing (6)
    • System performance as a function of calibration methods, sample size and sampling variability in likelihood ratio-based forensic voice comparison
      (3 minutes introduction)
    • System performance as a function of calibration methods, sample size and sampling variability in likelihood ratio-based forensic voice comparison
      (longer introduction)
    • The Four-way Classification of Stops with Voicing and Aspiration for Non-native Speech Evaluation
      (3 minutes introduction)
    • A comparison of the accuracy of Dissen and Keshets’s (2016) DeepFormants and traditional LPC methods for semi-automatic speaker recognition
      (3 minutes introduction)
    • Sound change in spontaneous bilingual speech: A corpus study on the Cantonese n-l merger in Cantonese-English bilinguals
      (3 minutes introduction)
    • Sound change in spontaneous bilingual speech: A corpus study on the Cantonese n-l merger in Cantonese-English bilinguals
      (longer introduction)
  • Voice Anti-Spoofing and Countermeasure (11)
    • A Comparative Study on Recent Neural Spoofing Countermeasures for Synthetic Speech Detection
      (3 minutes introduction)
    • An Initial Investigation for Detecting Partially Spoofed Audio
      (3 minutes introduction)
    • Cross-database replay detection in terminal-dependent speaker verification
      (3 minutes introduction)
    • Pairing Weak with Strong: Twin Models for Defending against Adversarial Attack on Speaker Verification
      (3 minutes introduction)
    • Pairing Weak with Strong: Twin Models for Defending against Adversarial Attack on Speaker Verification
      (longer introduction)
    • Voting for the right answer: Adversarial defense for speaker verification
      (3 minutes introduction)
    • Visualizing Classifier Adjacency Relations: A Case Study in Speaker Verification and Voice Anti-Spoofing
      (3 minutes introduction)
    • Representation Learning to Classify and Detect Adversarial Attacks against Speaker and Speech Recognition Systems
      (3 minutes introduction)
    • An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems
      (3 minutes introduction)
    • An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems
      (longer introduction)
    • Channel-wise Gated Res2Net: Towards Robust Detection of Synthetic Speech Attacks
      (3 minutes introduction)
  • Voice Conversion and Adaptation I (7)
    • CVC: Contrastive Learning for Non-parallel Voice Conversion
      (3 minutes introduction)
    • A Preliminary Study of a Two-Stage Paradigm for Preserving Speaker Identity in Dysarthric Voice Conversion
      (3 minutes introduction)
    • Fine-tuning pre-trained voice conversion model for adding new target speakers with limited data
      (3 minutes introduction)
    • Normalization Driven Zero-shot Multi-Speaker Speech Synthesis
      (3 minutes introduction)
    • StarGAN-VC+ASR: StarGAN-based Non-Parallel Voice Conversion Regularized by Automatic Speech Recognition
      (3 minutes introduction)
    • Two-Pathway Style Embedding for Arbitrary Voice Conversion
      (3 minutes introduction)
    • Non-Parallel Any-to-Many Voice Conversion by Replacing Speaker Statistics
      (3 minutes introduction)
  • Voice Conversion and Adaptation II (4)
    • Limited Data Emotional Voice Conversion Leveraging Text-to-Speech: Two-stage Sequence-to-Sequence Training
      (3 minutes introduction)
    • Adversarial Voice Conversion against Neural Spoofing Detectors
      (3 minutes introduction)
    • Adversarially Learning Disentangled Speech Representations for Robust Multi-factor Voice Conversion
      (3 minutes introduction)
    • Many-to-Many Voice Conversion based Feature Disentanglement using Variational Autoencoder
      (3 minutes introduction)
  • Voice quality characterization for clinical voice assessment: Voice production, acoustics, and auditory perception (4)
    • Optimizing an automatic creaky voice detection method for Australian English-speaking females
      (Oral presentation)
    • Investigating voice function characteristics of Greek speakers with hearing loss using automatic glottal source feature extraction
      (Oral presentation)
    • Automated Detection of Voice Disorder in the Saarbruecken Voice Database: Effects of Pathology Subset and Audio Materials
      (Oral presentation)
    • Articulatory Coordination for Speech Motor Tracking in Huntington Disease
      (Oral presentation)
  • Opening (1)
    • Opening ceremony
  • Closing (3)
    • Closing ceremony, awards
    • Closing ceremony, awards
    • Closing ceremony, awards

• Statistika