The task of record linkage aims to link records from diverse datasets into cohesive clusters, where each cluster represents a unique entity. The record linkage process is vital for applications such as healthcare, fraud detection, law enforcement, biology, transportation, and supply chain management. The sheer volume of data involved often poses significant computational challenges. To address these challenges, the blocking technique has emerged as a key strategy to streamline the record linkage process by reducing the number of computations. In this paper, we explore a novel application of Double Metaphone encoding, as a foundation for an innovative blocking method. Traditionally utilized for measuring phonetic similarity, Double Metaphone is here repurposed to enhance the blocking efficiency of the record linkage process. We introduce Double Metaphone as a blocking technique. Our research unveils improvements in record linkage outcomes, leveraging this new approach to achieve greater performance and reduced computational overhead. Through experiments, we demonstrate that our Double Metaphone Blocking approach outperforms the state-of-the-art record linkage algorithms as evidenced by reduced runtime and higher F-1 scores. This approach offers a promising solution for managing large scale data linkage tasks across various domains.