Log In Sign Up

Orthros: Non-autoregressive End-to-end Speech Translation with Dual-decoder

by   Hirofumi Inaguma, et al.

Fast inference speed is an important goal towards real-world deployment of speech translation (ST) systems. End-to-end (E2E) models based on the encoder-decoder architecture are more suitable for this goal than traditional cascaded systems, but their effectiveness regarding decoding speed has not been explored so far. Inspired by recent progress in non-autoregressive (NAR) methods in text-based translation, which generates target tokens in parallel by eliminating conditional dependencies, we study the problem of NAR decoding for E2E-ST. We propose a novel NAR E2E-ST framework, Orthoros, in which both NAR and autoregressive (AR) decoders are jointly trained on the shared speech encoder. The latter is used for selecting better translation among various length candidates generated from the former, which dramatically improves the effectiveness of a large length beam with negligible overhead. We further investigate effective length prediction methods from speech inputs and the impact of vocabulary sizes. Experiments on four benchmarks show the effectiveness of the proposed method in improving inference speed while maintaining competitive translation quality compared to state-of-the-art AR E2E-ST systems.


page 1

page 2

page 3

page 4


Non-autoregressive End-to-end Speech Translation with Parallel Autoregressive Rescoring

This article describes an efficient end-to-end speech translation (E2E-S...

High-Speed and High-Quality Text-to-Lip Generation

As a key component of talking face generation, lip movements generation ...

Fast-MD: Fast Multi-Decoder End-to-End Speech Translation with Non-Autoregressive Hidden Intermediates

The multi-decoder (MD) end-to-end speech translation model has demonstra...

Diffsound: Discrete Diffusion Model for Text-to-sound Generation

Generating sound effects that humans want is an important topic. However...

General-purpose, long-context autoregressive modeling with Perceiver AR

Real-world data is high-dimensional: a book, image, or musical performan...

End-to-end translation of human neural activity to speech with a dual-dual generative adversarial network

In a recent study of auditory evoked potential (AEP) based brain-compute...

Exploring Non-Autoregressive End-To-End Neural Modeling For English Mispronunciation Detection And Diagnosis

End-to-end (E2E) neural modeling has emerged as one predominant school o...