Alternating Weak Triphone/BPE Alignment Supervision from Hybrid Model Improves End-to-End ASR
This work addresses performance improvement for ASR systems, but it is incremental as it builds on existing hybrid models and regularization techniques.
The paper tackles the problem of improving end-to-end automatic speech recognition (ASR) model training by proposing alternating weak triphone/BPE alignment supervision, resulting in over 10% relative error rate reduction on the TED-LIUM 2 dataset compared to a baseline system.
In this paper, alternating weak triphone/BPE alignment supervision is proposed to improve end-to-end model training. Towards this end, triphone and BPE alignments are extracted using a pre-existing hybrid ASR system. Then, regularization effect is obtained by cross-entropy based intermediate auxiliary losses computed on such alignments at a mid-layer representation of the encoder for triphone alignments and at the encoder for BPE alignments. Weak supervision is achieved through strong label smoothing with parameter of 0.5. Experimental results on TED-LIUM 2 indicate that either triphone or BPE alignment based weak supervision improves ASR performance over standard CTC auxiliary loss. Moreover, their combination lowers the word error rate further. We also investigate the alternation of the two auxiliary tasks during model training, and additional performance gain is observed. Overall, the proposed techniques result in over 10% relative error rate reduction over a CTC-regularized baseline system.