To enhance the extraction of local features while preserving the contextual extraction ability of the Conformer encoder, an end-to-end speech recognition method called SEResConformer is proposed. This ...