Efficient Neural Architecture Search For End-to-end Speech Recognition Via Straight-through Gradients

Abstract

Neural Architecture Search (NAS), the process of automating architecture engineering, is an appealing next step to advancing end-to-end Automatic Speech Recognition (ASR), replacing expert-designed networks with learned, task-specific architectures. In contrast to early computational-demanding NAS methods, recent gradient-based NAS methods, e.g., DARTS (Differentiable ARchiTecture Search), SNAS (Stochastic NAS) and ProxylessNAS, significantly improve the NAS efficiency. In this paper, we make two contributions. First, we rigorously develop an efficient NAS method via Straight-Through (ST) gradients, called ST-NAS. Basically, ST-NAS uses the loss from SNAS but uses ST to back-propagate gradients through discrete variables to optimize the loss, which is not revealed in ProxylessNAS. Using ST gradients to support sub-graph sampling is a core element to achieve efficient NAS beyond DARTS and SNAS. Second, we successfully apply ST-NAS to end-to-end ASR. Experiments over the widely benchmark

Stats

• citations12
• S2 citations—
• github stars0
• HF likes0
• heat score8.35
• arxiv keyzheng2020efficient

Related papers

• Improved Conformer-based End-to-end Speech Recognition Using Neural Architecture Search (2021)0.00
• DARTS-ASR: Differentiable Architecture Search For Multilingual Speech Recognition And Adaptation (2020)8.60
• Latency-controlled Neural Architecture Search For Streaming Speech Recognition (2021)0.00
• Enhancing Speech Emotion Recognition Through Differentiable Architecture Search (2023)0.00
• Efficienttdnn: Efficient Architecture Search For Speaker Recognition (2021)10.07
• TSNAT: Two-step Non-autoregressvie Transformer Models For Speech Recognition (2021)10.61
• Effectiveasr: A Single-step Non-autoregressive Mandarin Speech Recognition Architecture With High Accuracy And Inference Speed (2024)3.58
• End-to-end Keyword Spotting Using Neural Architecture Search And Quantization (2021)8.60