Improved Long-Form Spoken Language Translation with Large Language Models
This work addresses the problem of high-quality translation for long-form spoken content, which is incremental as it builds on existing segmentation methods.
The paper tackles the challenge of translating long-form spoken language by fine-tuning a large language model to segment ASR transcripts for better translation quality, achieving an average improvement of 2.7 BLEU score across three languages compared to an automatic punctuation baseline.
A challenge in spoken language translation is that plenty of spoken content is long-form, but short units are necessary for obtaining high-quality translations. To address this mismatch, we fine-tune a general-purpose, large language model to split long ASR transcripts into segments that can be independently translated so as to maximize the overall translation quality. We compare to several segmentation strategies and find that our approach improves BLEU score on three languages by an average of 2.7 BLEU overall compared to an automatic punctuation baseline. Further, we demonstrate the effectiveness of two constrained decoding strategies to improve well-formedness of the model output from above 99% to 100%.