Intermediate Loss Regularization for CTC-based Speech Recognition
This work provides an incremental improvement for researchers and practitioners working on CTC-based ASR systems.
This paper introduces an intermediate CTC loss function to regularize CTC-based speech recognition models. This method achieved a 9.9% WER on the WSJ corpus and 5.2% CER on the AISHELL-1 corpus using greedy search without a language model.
We present a simple and efficient auxiliary loss function for automatic speech recognition (ASR) based on the connectionist temporal classification (CTC) objective. The proposed objective, an intermediate CTC loss, is attached to an intermediate layer in the CTC encoder network. This intermediate CTC loss well regularizes CTC training and improves the performance requiring only small modification of the code and small and no overhead during training and inference, respectively. In addition, we propose to combine this intermediate CTC loss with stochastic depth training, and apply this combination to a recently proposed Conformer network. We evaluate the proposed method on various corpora, reaching word error rate (WER) 9.9% on the WSJ corpus and character error rate (CER) 5.2% on the AISHELL-1 corpus respectively, based on CTC greedy search without a language model. Especially, the AISHELL-1 task is comparable to other state-of-the-art ASR systems based on auto-regressive decoder with beam search.