We demonstrate the potential of few-shot translation systems, trained with unpaired
language data, for both high and low-resource language pairs. We show that with only 5 …

Generative large language models (LLMs), eg, ChatGPT, have demonstrated remarkable
proficiency across several NLP tasks such as machine translation, question answering, text …

Large-scale automatic speech translation systems today lack key features that help machine-
mediated communication feel seamless when compared to human-to-human dialogue. In …

In the context of a more linked and globalized society, the significance of proficient cross-
cultural communication has been increasing to a position of utmost importance. Language …

Conformer-based models have become the dominant end-to-end architecture for speech
processing tasks. With the objective of enhancing the conformer architecture for efficient …

Direct speech-to-speech translation (S2ST), in which all components can be optimized
jointly, is advantageous over cascaded approaches to achieve fast inference with a …

We present Translatotron 2, a neural direct speech-to-speech translation model that can be
trained end-to-end. Translatotron 2 consists of a speech encoder, a linguistic decoder, an …

Abstract End-to-end (E2E) speech-to-text translation (ST) often depends on pretraining its
encoder and/or decoder using source transcripts via speech recognition or text translation …

This paper reports on the shared tasks organized by the 20th IWSLT Conference. The
shared tasks address 9 scientific challenges in spoken language translation: simultaneous …

The gap between speech and text modalities is a major challenge in speech-to-text
translation (ST). Different methods have been proposed to reduce this gap, but most of them …