Natural Language Processing
Information Extraction Engine
LcT has developed a software engine for information extraction from audio or text. Using natural language processing, the engine can extract the desired information such as who, when, where and what automatically. The engine will understand the meaning of the text and provide accurate extraction results. Based on the engine, many new applications or products can be developed. For example, the engine can support semantic search, summarization, information sorting, Q&A, and many new applications. In semantic search, instead of inputting keywords, searchers can input the desired search information as audio or text. The software engine consists of transcription, natural language processing, and translation. It leverages LcT's experience and expertise in noise reduction, speech enhancement, robust speech recognition, machine learning, language modeling, and natural language processing.
Available for Licensing
The engine is ready for licensing to enterprises who want to build more powerful and intelligent search engines to migrate from traditional keyword search to the next generation of semantic search or other applications. A Chinese version of the engine is operational and allows users to search Chinese text or messages and extract the desired information. The output can be translated to English with high accuracy.