Efficient Monotonic Multihead Attention Article Swipe

PDF

Related Concepts

Monotonic function Computer science Translation (biology) Inference Machine translation Task (project management) Variance (accounting) Artificial intelligence Natural language processing Reduction (mathematics) Algorithm Speech recognition Mathematics Economics Gene Geometry Messenger RNA Chemistry Management Business Mathematical analysis Biochemistry Accounting

Xutai Ma , Anna Sun , Siqi Ouyang , Hirofumi Inaguma , Paden Tomasello ·

YOU? · · 2023 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2312.04515 · OA: W4389500133

We introduce the Efficient Monotonic Multihead Attention (EMMA), a state-of-the-art simultaneous translation model with numerically-stable and unbiased monotonic alignment estimation. In addition, we present improved training and inference strategies, including simultaneous fine-tuning from an offline translation model and reduction of monotonic alignment variance. The experimental results demonstrate that the proposed model attains state-of-the-art performance in simultaneous speech-to-text translation on the Spanish and English translation task.