# Blending LLMs into Cascaded Speech Translation: KIT's Offline Speech Translation System for IWSLT 2024

arXiv: https://arxiv.org/abs/2406.16777 · HTML: https://mlsanity.com/paper/2406.16777/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 35 references that are in the map.
- First posted on arXiv 24 Jun 2024.

## Builds on (most cited first, 17 of 17)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [fairseq: A Fast, Extensible Toolkit for Sequence Modeling](https://mlsanity.com/paper/1904.01038/) (2019)
- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [Multilingual Denoising Pre-training for Neural Machine Translation](https://mlsanity.com/paper/2001.08210/) (2020)
- [Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation](https://mlsanity.com/paper/2004.09813/) (2020)
- [VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation](https://mlsanity.com/paper/2101.00390/) (2021)
- [TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptation](https://mlsanity.com/paper/1805.04699/) (2018)
- [GenTranslate: Large Language Models are Generative Multilingual Speech and Machine Translators](https://mlsanity.com/paper/2402.06894/) (2024)
- [Contextual Refinement of Translations: Large Language Models for Sentence and Document-Level Post-Editing](https://mlsanity.com/paper/2310.14855/) (2023)
- [Multi-stage Large Language Model Correction for Speech Recognition](https://mlsanity.com/paper/2310.11532/) (2023)
- [KIT's Multilingual Speech Translation System for IWSLT 2023](https://mlsanity.com/paper/2306.05320/) (2023)

## Built on by (most cited first, 4 of 4)

- [KIT's Offline Speech Translation and Instruction Following Submission for IWSLT 2025](https://mlsanity.com/paper/2505.13036/) (2025)
- [Speech Translation Refinement using Large Language Models](https://mlsanity.com/paper/2501.15090/) (2025)
- [Toward Machine Interpreting: Lessons from Human Interpreting Studies](https://mlsanity.com/paper/2508.07964/) (2025)
- [PART: Progressive Alignment Representation Training for Multilingual Speech-To-Text with LLMs](https://mlsanity.com/paper/2509.19745/) (2025)

## Abstract (arXiv)

Large Language Models (LLMs) are currently under exploration for various tasks, including Automatic Speech Recognition (ASR), Machine Translation (MT), and even End-to-End Speech Translation (ST). In this paper, we present KIT's offline submission in the constrained + LLM track by incorporating recently proposed techniques that can be added to any cascaded speech translation. Specifically, we integrate Mistral-7Bfootnotemistralai/Mistral-7B-Instruct-v0.1 into our system to enhance it in two ways. Firstly, we refine the ASR outputs by utilizing the N-best lists generated by our system and fine-tuning the LLM to predict the transcript accurately. Secondly, we refine the MT outputs at the document level by fine-tuning the LLM, leveraging both ASR and MT predictions to improve translation quality. We find that integrating the LLM into the ASR and MT systems results in an absolute improvement of 0.3% in Word Error Rate and 0.65% in COMET for tst2019 test set. In challenging test sets with overlapping speakers and background noise, we find that integrating LLM is not beneficial due to poor ASR performance. Here, we use ASR with chunked long-form decoding to improve context usage that may be unavailable when transcribing with Voice Activity Detection segmentation alone.