The technology that underlines the foundation of Heartsome's Translation tools is known as "translation memory" (TM). It is a language technology that enables the translation of segments (paragraphs, sentences or phrases) of documents by searching for similar segments in a database and suggesting matches that are found in the database as possible translations. It may further suggest several translations for a paragraph, a sentence or a phrase.
The translation memory application, which implements a search logic designed by the developer of the tool, is supported by a set of file format converters. These converters analyse the source file in such a manner that another tool, the segmenter, can come in to break the file into segments.
The database module, which serves as the core of Heartsome's tools, is specially designed to support unlimited languages across all operating systems. It comes complete with a cross-lingual search engine that automatically searches the translation memory database for matching entries for each and every segment. The level of match varies from 100% matches to a low of 35%. This facility enables users to locate, retrieve and reuse translations done previously, whether done several years ago or just a few minutes back. Without this facility, users may end up translating the same text repeatedly. Apart from duplicating the effort, it is also very likely that the same text may be translated differently, thus affecting the consistency in the document.
Heartsome's database engine is fully compliant with the one and only industry standard for managing and exchanging translation memory - TMX (Translation Memory eXchange). It is entirely possible to export databases generated or maintained by Heartsome translation tools for use in any other translation tools that support TMX Standard V1.4 or higher.
