A Chat About Boring Problems: Studying GPT-based text normalization
This addresses text normalization for spoken language processing, offering incremental improvements through LLM application and error analysis.
The paper tackled text normalization by demonstrating that large language models (LLMs) can achieve error rates around 40% lower than top systems in few-shot scenarios, and introduced a new taxonomy to analyze errors in GPT-based approaches.
Text normalization - the conversion of text from written to spoken form - is traditionally assumed to be an ill-formed task for language models. In this work, we argue otherwise. We empirically show the capacity of Large-Language Models (LLM) for text normalization in few-shot scenarios. Combining self-consistency reasoning with linguistic-informed prompt engineering, we find LLM based text normalization to achieve error rates around 40\% lower than top normalization systems. Further, upon error analysis, we note key limitations in the conventional design of text normalization tasks. We create a new taxonomy of text normalization errors and apply it to results from GPT-3.5-Turbo and GPT-4.0. Through this new framework, we can identify strengths and weaknesses of GPT-based TN, opening opportunities for future work.