Exploring the Correlation between Human and Machine Evaluation of Simultaneous Speech Translation
This addresses the challenge of reliable automated assessment for interpreting services, which is incremental as it applies existing methods to a specific domain.
This study tackled the problem of evaluating simultaneous speech translation by assessing the correlation between automatic metrics and human evaluations, finding that GPT-3.5 with direct prompting showed the strongest correlation with human judgment for semantic similarity, with context window size having a notable impact.
Assessing the performance of interpreting services is a complex task, given the nuanced nature of spoken language translation, the strategies that interpreters apply, and the diverse expectations of users. The complexity of this task become even more pronounced when automated evaluation methods are applied. This is particularly true because interpreted texts exhibit less linearity between the source and target languages due to the strategies employed by the interpreter. This study aims to assess the reliability of automatic metrics in evaluating simultaneous interpretations by analyzing their correlation with human evaluations. We focus on a particular feature of interpretation quality, namely translation accuracy or faithfulness. As a benchmark we use human assessments performed by language experts, and evaluate how well sentence embeddings and Large Language Models correlate with them. We quantify semantic similarity between the source and translated texts without relying on a reference translation. The results suggest GPT models, particularly GPT-3.5 with direct prompting, demonstrate the strongest correlation with human judgment in terms of semantic similarity between source and target texts, even when evaluating short textual segments. Additionally, the study reveals that the size of the context window has a notable impact on this correlation.