DeepAI AI Chat
Log In Sign Up

STACL: Simultaneous Translation with Integrated Anticipation and Controllable Latency

by   Mingbo Ma, et al.
Baidu, Inc.

Simultaneous translation, which translates sentences before they are finished, is useful in many scenarios but is notoriously difficult due to word-order differences and simultaneity requirements. We introduce a very simple yet surprisingly effective `wait-k' model trained to generate the target sentence concurrently with the source sentence, but always k words behind, for any given k. This framework seamlessly integrates anticipation and translation in a single model that involves only minor changes to the existing neural translation framework. Experiments on Chinese-to-English simultaneous translation achieve a 5-word latency with 3.4 (single-ref) BLEU points degradation in quality compared to full-sentence non-simultaneous translation. We also formulate a new latency metric that addresses deficiencies in previous ones.


page 1

page 2

page 3

page 4


Simultaneous Translation Policies: From Fixed to Adaptive

Adaptive policies are better than fixed policies for simultaneous transl...

Improving Simultaneous Translation with Pseudo References

Simultaneous translation is vastly different from full-sentence translat...

Full-Sentence Models Perform Better in Simultaneous Translation Using the Information Enhanced Decoding Strategy

Simultaneous translation, which starts translating each sentence after r...

Fluent and Low-latency Simultaneous Speech-to-Speech Translation with Self-adaptive Training

Simultaneous speech-to-speech translation is widely useful but extremely...

Opportunistic Decoding with Timely Correction for Simultaneous Translation

Simultaneous translation has many important application scenarios and at...

Does Simultaneous Speech Translation need Simultaneous Models?

In simultaneous speech translation (SimulST), finding the best trade-off...

Simultaneous Translation with Flexible Policy via Restricted Imitation Learning

Simultaneous translation is widely useful but remains one of the most di...