Combining Language Models For Specialized Domains: A Colorful Approach
This addresses a challenge for automatic speech recognition systems in specialized domains like medicine or industry, though it appears incremental as it builds on existing LM integration methods.
The paper tackles the problem of general-purpose language models struggling with domain-specific jargon and mixed speech by introducing a novel approach that integrates a domain-specific LM into a general-purpose LM using word labeling, which substantially lowers error rates for domain-specific words without compromising general performance.
General purpose language models (LMs) encounter difficulties when processing domain-specific jargon and terminology, which are frequently utilized in specialized fields such as medicine or industrial settings. Moreover, they often find it challenging to interpret mixed speech that blends general language with specialized jargon. This poses a challenge for automatic speech recognition systems operating within these specific domains. In this work, we introduce a novel approach that integrates domain-specific or secondary LM into general-purpose LM. This strategy involves labeling, or "coloring", each word to indicate its association with either the general or the domain-specific LM. We develop an optimized algorithm that enhances the beam search algorithm to effectively handle inferences involving colored words. Our evaluations indicate that this approach is highly effective in integrating jargon into language tasks. Notably, our method substantially lowers the error rate for domain-specific words without compromising performance in the general domain.