Interspeech 2021

979 papers

Top Research Directions

Research directions most represented by Interspeech 2021 papers.

Partitioning Variance with Linear Mixed Effects Models 34
Integrating Local and Global Context in Speech Architectures 31
Learning High-Level Acoustic Structures from Unlabeled Audio 30
The Single Speaker Generalization Bottleneck in Speech Synthesis 27
Balancing Latency and Accuracy in Streaming Speech Recognition 26
Zero Shot Speaker Generalization for Speech Synthesis 24
Overcoming Data Scarcity in Low-Resource Speech Recognition 23
Generalization to Unseen Acoustic Environments 20
Generalizing from Limited Cohorts in Speech Research 20
The Perceptual Evaluation Gap in Generative Audio 19
Resolving Overlapping Speech in Multi-Speaker Audio 18
Conformer Architectures for Perceptual Speech Enhancement 17
Bridging the Gap Between Read and Spontaneous Speech 17
Robust Natural Language Understanding for Noisy Speech Transcripts 17
Complex Spectral Mapping for Phase Aware Signal Reconstruction 16

Top Authors

AuthorPapers
Shinji Watanabe20
Haizhou Li12
Binghuai Lin10
Jinyu Li10
Jing Xiao9
Christian Fuegen8
Jun Du8
Brian Kingsbury7
Dong Yu7
Hung-yi Lee7
Najim Dehak7
Athanasios Mouchtaris6
Helen M. Meng6
James R. Glass6
Jesus Villalba6
Jianzong Wang6
Longbiao Wang6
Naoyuki Kanda6
Peter Bell6
Piotr Zelasko6

Impactful Papers

Papers from Interspeech 2021 that shaped research clusters.

Foundational

Influential

Papers

Title
Late Fusion of the Available Lexicon and Raw Waveform-Based Acoustic Modeling for Depression and Dementia Recognition
Layer Pruning on Demand with Intermediate CTC
Layer-Wise Fast Adaptation for End-to-End Multi-Accent Speech Recognition
LEAP Submission for the Third DIHARD Diarization Challenge
Learning a Neural Diff for Speech Models
Learning Explicit Prosody Models and Deep Speaker Embeddings for Atypical Voice Conversion
Learning Fine-Grained Cross Modality Excitement for Speech Emotion Recognition
Learning Mutual Correlation in Multimodal Transformer for Speech Emotion Recognition
Learning Robust Speech Representation with an Articulatory-Regularized Variational Autoencoder
Learning Speech Models from Multi-Modal Data
Learning Speech Structure to Improve Time-Frequency Masks
Learning to Rank Microphones for Distant Speech Recognition
LeBenchmark: A Reproducible Framework for Assessing Self-Supervised Representation Learning from Speech
Leveraging ASR N-Best in Deep Entity Retrieval
Leveraging Non-Target Language Resources to Improve ASR Performance in a Target Language
Leveraging Phone Mask Training for Phonetic-Reduction-Robust E2E Uyghur Speech Recognition
Leveraging Pre-Trained Language Model for Speech Sentiment Analysis
Leveraging Real-Time MRI for Illuminating Linguistic Velum Action
Leveraging Speaker Attribute Information Using Multi Task Learning for Speaker Verification and Diarization
Leveraging the Uniformity Framework to Examine Crosslinguistic Similarity for Long-Lag Stops in Spontaneous Cantonese-English Bilingual Speech
Lexical Density Analysis of Word Productions in Japanese English Using Acoustic Word Embeddings
Lexical Entrainment and Intra-Speaker Variability in Cooperative Dialogues
Lexical Modeling of ASR Errors for Robust Speech Translation
Librispeech Transducer Model with Internal Language Model Prior Correction
Lightweight Causal Transformer with Local Self-Attention for Real-Time Speech Enhancement
Limited Data Emotional Voice Conversion Leveraging Text-to-Speech: Two-Stage Sequence-to-Sequence Training
LinearSpeech: Parallel Text-to-Speech with Linear Complexity
LiRA: Learning Visual Speech Representations from Audio Through Self-Supervision
Listen with Intent: Improving Speech Recognition with Audio-to-Intent Front-End
LiteTTS: A Lightweight Mel-Spectrogram-Free Text-to-Wave Synthesizer Based on Generative Adversarial Networks
Live Subtitling for BigBlueButton with Open-Source Software
Live TV Subtitling Through Respeaking
Log-Likelihood-Ratio Cost Function as Objective Loss for Speaker Verification Systems
Lookup-Table Recurrent Language Models for Long Tail Speech Recognition
Look Who's Talking: Active Speaker Detection in the Wild
Lost in Interpreting: Speech Translation from Source or Interpreter?
Low-Delay Speech Enhancement Using Perceptually Motivated Target and Loss
Low Resource ASR: The Surprising Effectiveness of High Resource Transliteration
Low Resource German ASR with Untranscribed Data Spoken by Non-Native Children — INTERSPEECH 2021 Shared Task SPAPL System
LT-LM: A Novel Non-Autoregressive Language Model for Single-Shot Lattice Rescoring
M3: MultiModal Masking Applied to Sentiment Analysis
Manifold-Aware Deep Clustering: Maximizing Angles Between Embedding Vectors Based on Regular Simplex
Many-Speakers Single Channel Speech Separation with Optimal Permutation Training
Many-to-Many Voice Conversion Based Feature Disentanglement Using Variational Autoencoder
MAP Adaptation Characteristics in Forensic Long-Term Formant Analysis
Masked Proxy Loss for Text-Independent Speaker Verification
Measuring Voice Quality Parameters After Speaker Pseudonymization
MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement
Metric Learning Based Feature Representation with Gated Fusion Model for Speech Emotion Recognition
MetricNet: Towards Improved Modeling For Non-Intrusive Speech Quality Assessment

« prev Page 11/20 next »