SPEECH & AUDIO

NLP

Unveiling the Role of Pretraining in Direct Speech Translation

September 26, 2024

Abstract

Direct speech-to-text translation systems encounter an important drawback in data scarcity. A common solution consists on pretraining the encoder on automatic speech recognition, hence losing efficiency in the training process. In this study, we compare the training dynamics of a system using a pretrained encoder, the conventional approach, and one trained from scratch. We observe that, throughout the training, the randomly initialized model struggles to incorporate information from the speech inputs for its predictions. Hence, we hypothesize that this issue stems from the difficulty of effectively training an encoder for direct speech translation. While a model trained from scratch needs to learn acoustic and semantic modeling simultaneously, a pretrained one can just focus on the latter. Based on these findings, we propose a subtle change in the decoder cross-attention to integrate source information from earlier steps in training. We show that with this change, the model trained from scratch can achieve comparable performance to the pretrained one, while reducing the training time.

Download the Paper

AUTHORS

Written by

Belen Alastruey

Gerard I. Gállego

Marta R. Costa-jussa

Publisher

EMNLP

Related Publications

October 16, 2024

SPEECH & AUDIO

COMPUTER VISION

Movie Gen: A Cast of Media Foundation Models

Movie Gen Team

October 16, 2024

October 04, 2024

HUMAN & MACHINE INTELLIGENCE

CONVERSATIONAL AI

Beyond Turn-Based Interfaces: Synchronous LLMs as Full-Duplex Dialogue Agents

Bandhav Veluri, Benjamin Peloquin, Bokai Yu, Hongyu Gong, Shyam Gollakota

October 04, 2024

October 03, 2024

NLP

BLASER 2.0: a metric for evaluation and quality estimation of massively multilingual speech and text translation

David Dale, Marta R. Costa-jussa

October 03, 2024

September 05, 2024

CONVERSATIONAL AI

NLP

Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model

Chunting Zhou, Lili Yu, Arun Babu, Kushal Tirumala, Michihiro Yasunaga, Leonid Shamis, Jacob Kahn, Luke Zettlemoyer, Omer Levy, Xuezhe Ma

September 05, 2024

Help Us Pioneer The Future of AI

We share our open source frameworks, tools, libraries, and models for everything from research exploration to large-scale production deployment.