Knowledge Technology at KONVENS'25
12 September 2025
Our group members Theresa Pekarek-Rosin and Burak Can Kaplan presented their paper at the KONVENS'25 conference in Hildesheim, Germany. Here is more information about the paper:
Title: Large Language Model Data Generation for Enhanced Intent Recognition in German Speech
Authors: Theresa Pekarek-Rosin, Burak Can Kaplan, Stefan Wermter
Abstract: Intent recognition (IR) for speech commands is essential for artificial intelligence (AI) assistant systems; however, most existing approaches are limited to short commands and are predominantly developed for English. This paper addresses these limitations by focusing on IR from speech by elderly German speakers. We propose a novel approach that combines an adapted Whisper ASR model, fine-tuned on elderly German speech (SVC-de), with Transformer-based language models trained on synthetic text datasets generated by three well-known large language models (LLMs): LeoLM, Llama3, and ChatGPT. To evaluate the robustness of our approach, we generate synthetic speech with a text-to-speech model and conduct extensive cross-dataset testing. Our results show that synthetic LLM-generated data significantly boosts classification performance and robustness to different speaking styles and unseen vocabulary. Notably, we find that LeoLM, a smaller, domain-specific 13B LLM, surpasses the much larger ChatGPT (175B) in dataset quality for German intent recognition. Our approach demonstrates that generative AI can effectively bridge data gaps in low-resource domains. We provide detailed documentation of our data generation and training process to ensure transparency and reproducibility.
You can reach the full paper here.
In addition, at the closing session of the conference, it was officially announced that KONVENS 2026 will be hosted at our university.