Beyond Specialization: Benchmarking LLMs for Transliteration of Indian Languages

TL;DR

This study evaluates large language models' ability to perform transliteration of Indian languages, demonstrating that general-purpose LLMs can outperform specialized models in many cases, with potential for minimal fine-tuning.

Contribution

It systematically benchmarks prominent LLMs against a specialized transliteration model across multiple Indian languages, highlighting their strengths and robustness.

Findings

GPT models generally outperform specialized models like IndicXlit.

Fine-tuning GPT-4o enhances language-specific performance.

LLMs show robustness under noisy conditions.

Abstract

Transliteration, the process of mapping text from one script to another, plays a crucial role in multilingual natural language processing, especially within linguistically diverse contexts such as India. Despite significant advancements through specialized models like IndicXlit, recent developments in large language models suggest a potential for general-purpose models to excel at this task without explicit task-specific training. The current work systematically evaluates the performance of prominent LLMs, including GPT-4o, GPT-4.5, GPT-4.1, Gemma-3-27B-it, and Mistral-Large against IndicXlit, a state-of-the-art transliteration model, across ten major Indian languages. Experiments utilized standard benchmarks, including Dakshina and Aksharantar datasets, with performance assessed via Top-1 Accuracy and Character Error Rate. Our findings reveal that while GPT family models generally outperform other LLMs and IndicXlit for most instances. Additionally, fine-tuning GPT-4o improves performance on specific languages notably. An extensive error analysis and robustness testing under noisy conditions further elucidate strengths of LLMs compared to specialized models, highlighting the efficacy of foundational models for a wide spectrum of specialized applications with minimal overhead.