Thomas Pellegrini

convolutional neural network (4) transfer learning (3) data augmentation (2) automatic speech recognition (2) audio classification (2) phoneme recognition (2) self-supervised learning (1) speech transcription (1) logistic regression (1) text summarization (1) natural language processing (1) multitask learning (1) weak supervision (1) connectionist temporal classification (1) speaker diarization (1) weakly supervised learning (1) feature extraction (1) phone recognition (1) spectral clustering (1) feature learning (1)