Improving Generalization of Deep Neural Network Acoustic Models with Length Perturbation and N-best Based Label Smoothing

Autor:	Cui, Xiaodong, Saon, George, Nagano, Tohru, Suzuki, Masayuki, Fukuda, Takashi, Kingsbury, Brian, Kurata, Gakuto
Rok vydání:	2022
Předmět:	Computer Science - Computation and Language Computer Science - Sound Electrical Engineering and Systems Science - Audio and Speech Processing
Druh dokumentu:	Working Paper
Popis:	We introduce two techniques, length perturbation and n-best based label smoothing, to improve generalization of deep neural network (DNN) acoustic models for automatic speech recognition (ASR). Length perturbation is a data augmentation algorithm that randomly drops and inserts frames of an utterance to alter the length of the speech feature sequence. N-best based label smoothing randomly injects noise to ground truth labels during training in order to avoid overfitting, where the noisy labels are generated from n-best hypotheses. We evaluate these two techniques extensively on the 300-hour Switchboard (SWB300) dataset and an in-house 500-hour Japanese (JPN500) dataset using recurrent neural network transducer (RNNT) acoustic models for ASR. We show that both techniques improve the generalization of RNNT models individually and they can also be complementary. In particular, they yield good improvements over a strong SWB300 baseline and give state-of-art performance on SWB300 using RNNT models. Comment: Submitted to Interspeech 2022
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2203.15176 Zobrazit plný text záznamu View this record from Arxiv