# Improving speech recognition by revising gated recurrent units

arXiv: https://arxiv.org/abs/1710.00641 · HTML: https://mlsanity.com/paper/1710.00641/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Sep 2017.

## Built on by (most cited first, 3 of 3)

- [Shapes of Emotions: Multimodal Emotion Recognition in Conversations via Emotion Shifts](https://mlsanity.com/paper/2112.01938/) (2021)
- [Delayed Memory Unit: Modelling Temporal Dependency Through Delay Gate](https://mlsanity.com/paper/2310.14982/) (2023)
- [Personalized Event Prediction for Electronic Health Records](https://mlsanity.com/paper/2308.11013/) (2023)

## Abstract (arXiv)

Speech recognition is largely taking advantage of deep learning, showing that substantial benefits can be obtained by modern Recurrent Neural Networks (RNNs). The most popular RNNs are Long Short-Term Memory (LSTMs), which typically reach state-of-the-art performance in many tasks thanks to their ability to learn long-term dependencies and robustness to vanishing gradients. Nevertheless, LSTMs have a rather complex design with three multiplicative gates, that might impair their efficient implementation. An attempt to simplify LSTMs has recently led to Gated Recurrent Units (GRUs), which are based on just two multiplicative gates. This paper builds on these efforts by further revising GRUs and proposing a simplified architecture potentially more suitable for speech recognition. The contribution of this work is two-fold. First, we suggest to remove the reset gate in the GRU design, resulting in a more efficient single-gate architecture. Second, we propose to replace tanh with ReLU activations in the state update equations. Results show that, in our implementation, the revised architecture reduces the per-epoch training time with more than 30% and consistently improves recognition performance across different tasks, input features, and noisy conditions when compared to a standard GRU.