Automatic Speech Recognition (ASR) increasingly mediates access to broadcast media, public discourse and cultural archives. For minoritised languages, however, the development of robust ASR systems is constrained by limited and domain-restricted text data. This paper investigates cross-lingual text expansion (XLTE), a method that uses a Large Language Model (LLM) to generate in-domain text in a low-resource language from high-resource language summaries. We further examine whether supervised fine-tuning on a small set of human-authored texts enhances generation quality. Using Scottish Gaelic as a case study, we show that synthetic text generated via fine-tuned XLTE can be used to train an external language model that reduces Word Error Rate (WER) by 24.48% in a previously unseen broadcast domain. Our findings demonstrate that text-only domain adaptation through cross-lingual generation can strengthen speech technology in sparse data settings. Beyond engineering gains, the approach offers a scalable pathway for improving the digital representation, accessibility and sustainability of minoritised-language media and cultural heritage.

Large language models (LLMs) have shown significant potential for robotics applications, particularly task planning, by harnessing their language comprehension and text generation capabilities. However, in applications such as household robotics, a critical gap remains in the personalization of these models to household preferences. For example, an LLM planner may find it challenging to perform tasks that require personalization, such as deciding where to place mugs in a kitchen based on specific household preferences. We introduce LLM-Personalize, a novel framework designed to personalize LLM planners for household robotics. LLM-Personalize uses an LLM planner to perform iterative planning in multi-room, partially-observable household environments, utilizing a scene graph built dynamically from local observations. To personalize the LLM planner towards user preferences, our optimization pipeline integrates imitation learning and reinforced Self-Training. We evaluate LLM-Personalize on Housekeep, a challenging simulated real-world 3D benchmark for household rearrangements, demonstrating a more than 30 percent increase in success rate over existing LLM planners, showcasing significantly improved alignment with human preferences.

Advances in large language modelling have disproportionately benefited high-resource languages due to their vastly greater training data reserves. This paper proposes a novel cross-lingual text expansion (XLTE) technique using multilingual large language models (MLLMs) to mitigate data sparsity in low-resource languages. We apply XLTE to the domain of traditional Scottish Gaelic storytelling to generate a training corpus suitable for language modelling, for example as part of an automatic speech recognition system. The effectiveness of this technique is demonstrated using OpenAI’s GPT-4o, with supervised fine-tuning (SFT) providing decreased neologism rates and a 57.2% reduction in perplexity over the baseline model. Despite these promising results, qualitative analyses reveal important stylistic divergences between synthesised and genuine data. Nevertheless, XLTE offers a promising, scalable method for synthesising training sets in other languages and domains, opening avenues for further improvements in low-resource language modelling.