Monotonic Multihead Attention

by   Xutai Ma, et al.

Simultaneous machine translation models start generating a target sequence before they have encoded or read the source sequence. Recent approaches for this task either apply a fixed policy on a state-of-the art Transformer model, or a learnable monotonic attention on a weaker recurrent neural network-based structure. In this paper, we propose a new attention mechanism, Monotonic Multihead Attention (MMA), which extends the monotonic attention mechanism to multihead attention. We also introduce two novel and interpretable approaches for latency control that are specifically designed for multiple attentions heads. We apply MMA to the simultaneous machine translation task and demonstrate better latency-quality tradeoffs compared to MILk, the previous state-of-the-art approach. We also analyze how the latency controls affect the attention span and we motivate the introduction of our model by analyzing the effect of the number of decoder layers and heads on quality and latency.


page 1

page 2

page 3

page 4


Monotonic Infinite Lookback Attention for Simultaneous Machine Translation

Simultaneous machine translation begins to translate each source sentenc...

Infusing Future Information into Monotonic Attention Through Language Models

Simultaneous neural machine translation(SNMT) models start emitting the ...

The USTC-NELSLIP Systems for Simultaneous Speech Translation Task at IWSLT 2021

This paper describes USTC-NELSLIP's submissions to the IWSLT2021 Simulta...

Tree-to-tree Neural Networks for Program Translation

Program translation is an important tool to migrate legacy code in one l...

Universal Simultaneous Machine Translation with Mixture-of-Experts Wait-k Policy

Simultaneous machine translation (SiMT) generates translation before rea...

Gaussian Multi-head Attention for Simultaneous Machine Translation

Simultaneous machine translation (SiMT) outputs translation while receiv...

Neural HMMs are all you need (for high-quality attention-free TTS)

Neural sequence-to-sequence TTS has achieved significantly better output...