# Improved Conformer-based End-to-End Speech Recognition Using Neural Architecture Search

arXiv: https://arxiv.org/abs/2104.05390 · HTML: https://mlsanity.com/paper/2104.05390/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Apr 2021.

## Built on by (most cited first, 4 of 4)

- [Decoupling recognition and transcription in Mandarin ASR](https://mlsanity.com/paper/2108.01129/) (2021)
- [Neural Architecture Search for Speech Emotion Recognition](https://mlsanity.com/paper/2203.16928/) (2022)
- [A Comparative Study of Pretrained Transformer Models for Quranic ASR: Speech Representations, Label Formats, and Dataset Composition](https://mlsanity.com/paper/2606.19747/) (2026)
- [Dynamic Acoustic Model Architecture Optimization in Training for ASR](https://mlsanity.com/paper/2506.13180/) (2025)

## Abstract (arXiv)

Recently neural architecture search(NAS) has been successfully used in image classification, natural language processing, and automatic speech recognition(ASR) tasks for finding the state-of-the-art(SOTA) architectures than those human-designed architectures. NAS can derive a SOTA and data-specific architecture over validation data from a pre-defined search space with a search algorithm. Inspired by the success of NAS in ASR tasks, we propose a NAS-based ASR framework containing one search space and one differentiable search algorithm called Differentiable Architecture Search(DARTS). Our search space follows the convolution-augmented transformer(Conformer) backbone, which is a more expressive ASR architecture than those used in existing NAS-based ASR frameworks. To improve the performance of our method, a regulation method called Dynamic Search Schedule(DSS) is employed. On a widely used Mandarin benchmark AISHELL-1, our best-searched architecture outperforms the baseline Conform model significantly with about 11% CER relative improvement, and our method is proved to be pretty efficient by the search cost comparisons.