Aligning Pre-trained Models for Spoken Language Translation

TL;DR

This paper presents a scalable method for end-to-end spoken language translation by aligning pre-trained ASR and MT models with a small connector module, improving translation performance and domain adaptation.

Contribution

Introducing a novel alignment approach using a small connector module to bridge pre-trained ASR and MT models for speech translation.

Findings

Increasing foundation model sizes improves translation quality.

Connectors can serve as effective domain adapters.

The approach is scalable and maintains small connector size.

Abstract

This paper investigates a novel approach to end-to-end speech translation (ST) based on aligning frozen pre-trained automatic speech recognition (ASR) and machine translation (MT) models via a small connector module (Q-Former, our Subsampler-Transformer Encoder). This connector bridges the gap between the speech and text modalities, transforming ASR encoder embeddings into the latent representation space of the MT encoder while being the only part of the system optimized during training. Experiments are conducted on the How2 English-Portuguese dataset as we investigate the alignment approach in a small-scale scenario focusing on ST. While keeping the size of the connector module constant and small in comparison ( < 5% of the size of the larger aligned models), increasing the size and capability of the foundation ASR and MT models universally improves translation results. We also find that the connectors can serve as domain adapters for the foundation MT models, significantly improving translation performance in the aligned ST setting. We conclude that this approach represents a viable and scalable approach to training end-to-end ST systems.