New algorithms and approaches for speech recognition

USING MODIFIED ADULT SPEECH AS DATA AUGMENTATION FOR CHILD SPEECH RECOGNITION

M-SpeechCLIP: Leveraging Large-Scale, Pre-Trained Models for Multilingual Speech to Image Retrieval

FIXED-POINT QUANTIZATION AWARE TRAINING FOR ON-DEVICE KEYWORD-SPOTTING

SPEECH-TEXT BASED MULTI-MODAL TRAINING WITH BIDIRECTIONAL ATTENTION FOR IMPROVED SPEECH RECOGNITION

Learning Cross-lingual Visual Speech Representations

IMPROVING FAIRNESS AND ROBUSTNESS IN END-TO-END SPEECH RECOGNITION THROUGH UNSUPERVISED CLUSTERING

Cleanformer: A Multichannel Array Configuration-Invariant Neural Enhancement Frontend for ASR in Smart Speakers

IMPORTANCE OF DIFFERENT TEMPORAL MODULATIONS OF SPEECH: A TALE OF TWO PERSPECTIVES

Structured Pruning of Self-Supervised Pre-trained Models for Speech Recognition and Understanding

Group Personalized Federated Learning

token2vec: A Joint Self-Supervised Pre-training Framework Using Unpaired Speech and Text

Robust Knowledge Distillation from RNN-T Models With Noisy Training Labels Using Full-Sum Loss