Recently Convolution-augmented Transformer (Conformer) has shown promising
results in Automatic Speech Recognition (ASR), outperforming the previous best
published Transformer Transducer. In this work, we believe that the output
information of each block in the encoder and decoder is not completely
inclusive, in other words, their output information may be complementary. We
study how to take advantage of the complementary information of each block in a
parameter-efficient way, and it is expected that this may lead to more robust
performance. Therefore we propose the Block-augmented Transformer for speech
recognition, named Blockformer. We have implemented two block ensemble methods:
the base Weighted Sum of the Blocks Output (Base-WSBO), and the
Squeeze-and-Excitation module to Weighted Sum of the Blocks Output (SE-WSBO).
Experiments have proved that the Blockformer significantly outperforms the
state-of-the-art Conformer-based models on AISHELL-1, our model achieves a CER
of 4.29\% without using a language model and 4.05\% with an external language
model on the testset.

本文提出了一种基于 Conformer 的 Block-augmented Transformer 模型来更好地利用编码器和解码器中各块的互补信息，这个模型在 AISHELL-1 语音识别任务中性能优于当前的最新模型 Conformer，并且在测试集上取得了 4.29% CER（不使用语言模型）和 4.05% CER（使用外部语言模型）。