A translation memory (TM) is a database that stores "segments", which can be sentences, paragraphs or sentence-like models (headings, titles or elements in a listing) that have previously been translated, so as to help human translators. The translation memory shops the source text and Memory Wave Audio its corresponding translation in language pairs known as "translation items". Individual phrases are dealt with by terminology bases and should not inside the domain of TM. Software program programs that use translation reminiscences are generally often called translation memory managers (TMM) or translation memory programs (TM systems, not to be confused with a translation administration system (TMS), which is one other type of software program centered on managing the process of translation). Translation recollections are usually used along side a devoted pc-assisted translation (CAT) device, word processing program, terminology management methods, multilingual dictionary, and even raw machine translation output. Analysis indicates that many corporations producing multilingual documentation are using translation Memory Wave programs. In a survey of language professionals in 2006, 82.5% out of 874 replies confirmed using a TM.

Utilization of TM correlated with text sort characterised by technical terms and simple sentence structure (technical, to a lesser degree advertising and marketing and financial), computing skills, Memory Wave and repetitiveness of content material. This system breaks the supply textual content (the textual content to be translated) into segments, seems to be for matches between segments and the supply half of previously translated supply-goal pairs stored in a translation memory, and presents such matching pairs as translation full and partial matches. The translator can accept a match, replace it with a recent translation, or modify it to match the supply. Within the last two instances, the brand new or modified translation goes into the database. Some translation memory programs search for 100% matches only, i.e. they will solely retrieve segments of textual content that match entries in the database exactly, while others employ fuzzy matching algorithms to retrieve similar segments, which are offered to the translator with variations flagged. Typical translation memory programs only seek for text in the source segment. The pliability and robustness of the matching algorithm largely decide the efficiency of the translation memory, although for some functions the recall rate of exact matches can be high enough to justify the 100%-match strategy.

Segments the place no match is discovered must be translated by the translator manually. These newly translated segments are stored within the database where they can be utilized for future translations as well as repetitions of that section in the current textual content. Translation reminiscences work greatest on texts which are extremely repetitive, comparable to technical manuals. They are also helpful for translating incremental modifications in a previously translated document, corresponding, for example, to minor changes in a brand new model of a consumer manual. Traditionally, translation memories have not been thought-about appropriate for literary or creative texts, for the simple cause that there is so little repetition within the language used. Nonetheless, others discover them of worth even for non-repetitive texts, as a result of the database resources created have value for concordance searches to find out appropriate usage of terms, for quality assurance (no empty segments), and the simplification of the review process (source and target phase are at all times displayed together whereas translators must work with two paperwork in a conventional overview setting).

Translation Memory Wave Audio managers are most suitable for translating technical documentation and documents containing specialized vocabularies. Guaranteeing that the translated documents are constant, together with widespread definitions, phrasings and terminology. That is important when totally different translators are engaged on a single challenge. Enabling translators to translate documents in a wide variety of codecs without having to personal the software program sometimes required to course of these formats. Accelerating the overall translation process; since translation reminiscences "remember" beforehand translated materials, translators have to translate it only as soon as. Decreasing costs of long-time period translation tasks; for example the textual content of manuals, warning messages or series of documents must be translated solely once and can be used a number of occasions. For big documentation tasks, financial savings (in time or money) due to using a TM package deal may already be apparent even for the first translation of a brand new challenge, but usually such savings are only apparent when translating subsequent variations of a undertaking that was translated before utilizing translation memory.

The idea of "translation memories" is based on the premise that sentences utilized in earlier translations might be "recycled". However, a guiding principle of translation is that the translator should translate the message of the text, and not its element sentences. Translation memory managers don't simply fit into existing translation or localization processes. With a view to reap the benefits of TM technology, the translation processes should be redesigned. Translation memory managers do not presently support all documentation formats, and filters could not exist to assist all file varieties. There's a studying curve associated with using translation memory managers, and the packages should be custom-made for best effectiveness. In instances the place all or part of the translation process is outsourced or handled by freelance translators working off-site, the off-site workers require special instruments to be able to work with the texts generated by the translation memory supervisor. Full variations of many translation memory managers can price from US$500 to US$2,500 per seat, which might represent a substantial investment (although lower cost packages are also available).