Findings of the IWSLT 2024 Evaluation Campaign

This paper reports on the shared tasks organized by the 21st IWSLT Conference. The shared tasks address 7 scientific challenges in spoken language translation: simultaneous and offline translation, automatic subtitling and dubbing, speech-to-speech translation, dialect and low-resource speech translation, and Indic languages. The shared tasks attracted 18 teams whose submissions are documented in 26 system papers. The growing interest towards spoken language translation is also witnessed by the constantly increasing number of shared task organizers and contributors to the overview paper, almost evenly distributed across industry and academia.

A Word Order Synchronization Metric for Evaluating Simultaneous Interpretation and Translation

Simultaneous interpretation (SI), the translation of one language to another in real time, starts translation before the original speech has finished. Its evaluation needs to consider both latency and quality. This trade-off is challenging especially for distant word order language pairs such as English and Japanese. To handle this word order gap, interpreters maintain the word order of the source language as much as possible to keep up with original language to minimize its latency while maintaining its quality, whereas in translation reordering happens to keep fluency in the target language. This means outputs synchronized with the source language are desirable based on the real SI situation, and it's a key for further progress in computational SI and simultaneous machine translation (SiMT). In this work, we propose an automatic evaluation metric for SI and SiMT focusing on word order synchronization. Our evaluation metric is based on rank correlation coefficients, leveraging cross-lingual pre-trained language models. Our experimental results on NAIST-SIC-Aligned and JNPC showed our metrics' effectiveness to measure word order synchronization between source and target language.

NAIST Simultaneous Speech Translation System for IWSLT 2024

This paper describes NAIST's submission to the simultaneous track of the IWSLT 2024 Evaluation Campaign: English-to-{German, Japanese, Chinese} speech-to-text translation and English-to-Japanese speech-to-speech translation. We develop a multilingual end-to-end speech-to-text translation model combining two pre-trained language models, HuBERT and mBART. We trained this model with two decoding policies, Local Agreement (LA) and AlignAtt. The submitted models employ the LA policy because it outperformed the AlignAtt policy in previous models. Our speech-to-speech translation method is a cascade of the above speech-to-text model and an incremental text-to-speech (TTS) module that incorporates a phoneme estimation model, a parallel acoustic model, and a parallel WaveGAN vocoder. We improved our incremental TTS by applying the Transformer architecture with the AlignAtt policy for the estimation model. The results show that our upgraded TTS module contributed to improving the system performance.

LLMs Are Zero-Shot Context-Aware Simultaneous Translators

The advent of transformers has fueled progress in machine translation. More recently large language models (LLMs) have come to the spotlight thanks to their generality and strong performance in a wide range of language tasks, including translation. Here we show that open-source LLMs perform on par with or better than some state-of-the-art baselines in simultaneous machine translation (SiMT) tasks, zero-shot. We also demonstrate that injection of minimal background information, which is easy with an LLM, brings further performance gains, especially on challenging technical subject-matter. This highlights LLMs' potential for building next generation of massively multilingual, context-aware and terminologically accurate SiMT systems that require no resource-intensive training or fine-tuning.

Automated Essay Scoring Using Grammatical Variety and Errors with Multi-Task Learning and Item Response Theory

This study examines the effect of grammatical features in automatic essay scoring (AES). We use two kinds of grammatical features as input to an AES model: (1) grammatical items that writers used correctly in essays, and (2) the number of grammatical errors. Experimental results show that grammatical features improve the performance of AES models that predict the holistic scores of essays. Multi-task learning with the holistic and grammar scores, alongside using grammatical features, resulted in a larger improvement in model performance. We also show that a model using grammar abilities estimated using Item Response Theory (IRT) as the labels for the auxiliary task achieved comparable performance to when we used grammar scores assigned by human raters. In addition, we weight the grammatical features using IRT to consider the difficulty of grammatical items and writers' grammar abilities. We found that weighting grammatical features with the difficulty led to further improvement in performance.

Word Order in English-Japanese Simultaneous Interpretation: Analyses and Evaluation using Chunk-wise Monotonic Translation

This paper analyzes the features of monotonic translations, which follow the word order of the source language, in simultaneous interpreting (SI). The word order differences are one of the biggest challenges in SI, especially for language pairs with significant structural differences like English and Japanese. We analyzed the characteristics of monotonic translations using the NAIST English-to-Japanese Chunk-wise Monotonic Translation Evaluation Dataset and found some grammatical structures that make monotonic translation difficult in English-Japanese SI. We further investigated the features of monotonic translations through evaluating the output from the existing speech translation (ST) and simultaneous speech translation (simulST) models on NAIST English-to-Japanese Chunk-wise Monotonic Translation Evaluation Dataset as well as on existing test sets. The results suggest that the existing SI-based test set underestimates the model performance. We also found that the monotonic-translation-based dataset would better evaluate simulST models, while using an offline-based test set for evaluating simulST models underestimates the model performance.

Evaluating the IWSLT2023 Speech Translation Tasks: Human Annotations, Automatic Metrics, and Segmentation

Human evaluation is a critical component in machine translation system development and has received much attention in text translation research. However, little prior work exists on the topic of human evaluation for speech translation, which adds additional challenges such as noisy data and segmentation mismatches. We take first steps to fill this gap by conducting a comprehensive human evaluation of the results of several shared tasks from the last International Workshop on Spoken Language Translation (IWSLT 2023). We propose an effective evaluation strategy based on automatic resegmentation and direct assessment with segment context. Our analysis revealed that: 1) the proposed evaluation strategy is robust and scores well-correlated with other types of human judgements; 2) automatic metrics are usually, but not always, well-correlated with direct assessment scores; and 3) COMET as a slightly stronger automatic metric than chrF, despite the segmentation noise introduced by the resegmentation step systems. We release the collected human-annotated data in order to encourage further investigation.

TransLLaMa: LLM-based Simultaneous Translation System

Decoder-only large language models (LLMs) have recently demonstrated impressive capabilities in text generation and reasoning. Nonetheless, they have limited applications in simultaneous machine translation (SiMT), currently dominated by encoder-decoder transformers. This study demonstrates that, after fine-tuning on a small dataset comprising causally aligned source and target sentence pairs, a pre-trained open-source LLM can control input segmentation directly by generating a special "wait" token. This obviates the need for a separate policy and enables the LLM to perform English-German and English-Russian SiMT tasks with BLEU scores that are comparable to those of specific state-of-the-art baselines. We also evaluated closed-source models such as GPT-4, which displayed encouraging results in performing the SiMT task without prior training (zero-shot), indicating a promising avenue for enhancing future SiMT systems.

Average Token Delay: A Duration-aware Latency Metric for Simultaneous Translation

Simultaneous translation is a task in which the translation begins before the end of an input speech segment. Its evaluation should be conducted based on latency in addition to quality, and for users, the smallest possible amount of latency is preferable. Most existing metrics measure latency based on the start timings of partial translations and ignore their duration. This means such metrics do not penalize the latency caused by long translation output, which delays the comprehension of users and subsequent translations. In this work, we propose a novel latency evaluation metric for simultaneous translation called \emph{Average Token Delay} (ATD) that focuses on the duration of partial translations. We demonstrate its effectiveness through analyses simulating user-side latency based on Ear-Voice Span (EVS). In our experiment, ATD had the highest correlation with EVS among baseline latency metrics under most conditions.

Tagged End-to-End Simultaneous Speech Translation Training using Simultaneous Interpretation Data

Simultaneous speech translation (SimulST) translates partial speech inputs incrementally. Although the monotonic correspondence between input and output is preferable for smaller latency, it is not the case for distant language pairs such as English and Japanese. A prospective approach to this problem is to mimic simultaneous interpretation (SI) using SI data to train a SimulST model. However, the size of such SI data is limited, so the SI data should be used together with ordinary bilingual data whose translations are given in offline. In this paper, we propose an effective way to train a SimulST model using mixed data of SI and offline. The proposed method trains a single model using the mixed data with style tags that tell the model to generate SI- or offline-style outputs. Experiment results show improvements of BLEURT in different latency ranges, and our analyses revealed the proposed model generates SI-style outputs more than the baseline.