Ivan Sorokin

automatic speech recognition (3) acoustic model (3) speaker adaptation (2) room impulse response (2) speaker diarization (1) mixup training (1) word error rate (1) distant speech recognition (1) overlapping speech (1) diarization error rate (1) lattice-free mmi (1) neural network language model (1) multi-speaker detection (1) target-speaker voice activity detection (1) binary classification (1) r-vector embedding (1) embedding learning (1) self-supervised learning (1) data augmentation (1)