# Darts-Conformer: Towards Efficient Gradient-Based Neural Architecture Search For End-to-End ASR

arXiv: https://arxiv.org/abs/2104.02868 · HTML: https://mlsanity.com/paper/2104.02868/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Apr 2021.

## Built on by (most cited first, 10 of 10)

- [A Survey of Techniques for Optimizing Transformer Inference](https://mlsanity.com/paper/2307.07982/) (2023)
- [A Survey on Multimodal Wearable Sensor-based Human Action Recognition](https://mlsanity.com/paper/2404.15349/) (2024)
- [Training-free Neural Architecture Search for RNNs and Transformers](https://mlsanity.com/paper/2306.00288/) (2023)
- [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)
- [Hyper-parameter Adaptation of Conformer ASR Systems for Elderly and Dysarthric Speech Recognition](https://mlsanity.com/paper/2306.15265/) (2023)
- [Evolutionary Neural Architecture Search for Transformer in Knowledge Tracing](https://mlsanity.com/paper/2310.01180/) (2023)
- [Self-Programming Artificial Intelligence Using Code-Generating Language Models](https://mlsanity.com/paper/2205.00167/) (2022)
- [ASR Bundestag: A Large-Scale political debate dataset in German](https://mlsanity.com/paper/2302.06008/) (2023)
- [Breaking Robustness Barriers in Cognitive Diagnosis: A One-Shot Neural Architecture Search Perspective](https://mlsanity.com/paper/2601.04918/) (2026)

## Abstract (arXiv)

Neural architecture search (NAS) has been successfully applied to tasks like image classification and language modeling for finding efficient high-performance network architectures. In ASR field especially end-to-end ASR, the related research is still in its infancy. In this work, we focus on applying NAS on the most popular manually designed model: Conformer, and then propose an efficient ASR model searching method that benefits from the natural advantage of differentiable architecture search (Darts) in reducing computational overheads. We fuse Darts mutator and Conformer blocks to form a complete search space, within which a modified architecture called Darts-Conformer cell is found automatically. The entire searching process on AISHELL-1 dataset costs only 0.7 GPU days. Replacing the Conformer encoder by stacking searched cell, we get an end-to-end ASR model (named as Darts-Conformner) that outperforms the Conformer baseline by 4.7% on the open-source AISHELL-1 dataset. Besides, we verify the transferability of the architecture searched on a small dataset to a larger 2k-hour dataset. To the best of our knowledge, this is the first successful attempt to apply gradient-based architecture search in the attention-based encoder-decoder ASR model.