Log In Sign Up

Non-Autoregressive Transformer ASR with CTC-Enhanced Decoder Input

by   Xingchen Song, et al.

Non-autoregressive (NAR) transformer models have achieved significantly inference speedup but at the cost of inferior accuracy compared to autoregressive (AR) models in automatic speech recognition (ASR). Most of the NAR transformers take a fixed-length sequence filled with MASK tokens or a redundant sequence copied from encoder states as decoder input, they cannot provide efficient target-side information thus leading to accuracy degradation. To address this problem, we propose a CTC-enhanced NAR transformer, which generates target sequence by refining predictions of the CTC module. Experimental results show that our method outperforms all previous NAR counterparts and achieves 50x faster decoding speed than a strong AR baseline with only 0.0   0.3 absolute CER degradation on Aishell-1 and Aishell-2 datasets.


page 1

page 2

page 3

page 4


Improved Mask-CTC for Non-Autoregressive End-to-End ASR

For real-world deployment of automatic speech recognition (ASR), the sys...

FastLR: Non-Autoregressive Lipreading Model with Integrate-and-Fire

Lipreading is an impressive technique and there has been a definite impr...

Spike-Triggered Non-Autoregressive Transformer for End-to-End Speech Recognition

Non-autoregressive transformer models have achieved extremely fast infer...

Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition

Transformers have recently dominated the ASR field. Although able to yie...

TSNAT: Two-Step Non-Autoregressvie Transformer Models for Speech Recognition

The autoregressive (AR) models, such as attention-based encoder-decoder ...

Non-Autoregressive Transformer Automatic Speech Recognition

Recently very deep transformers start showing outperformed performance t...

Align-Refine: Non-Autoregressive Speech Recognition via Iterative Realignment

Non-autoregressive models greatly improve decoding speed over typical se...