SPEECH RECOGNITION APPARATUS CAPABLE OF OPTIMIZING A DOMAIN-BASED SPEECH RECOGNITION MODEL AND OPERATING METHOD THEREOF

摘要

Disclosed is a speech recognition apparatus capable of optimizing a domain-based speech recognition model and a method of operating the same. The present invention finds a combination of a voice recognition model and a user-defined word table suitable for a specific field among pre-built voice recognition models and pre-built user-defined word tables according to characteristics of each field (domain). Determining a speech recognition model specialized for the specific field, and providing a speech recognition apparatus and an operation method for performing speech recognition processing for the specific field based on the determined speech recognition model, thereby improving the speech recognition rate for the specific field You can plan.