There are more and better mobile translation applications, as well as agile, reliable and precise services to adapt content from one language to another.
Automatic translation is the name given to an area of the interdisciplinary field known as computational linguistics that focuses on software intended to adapt, from one natural language to another, voice and/or text content.
These systems basically replace words from a certain language with their corresponding pairs from another language. For this reason, sometimes the results are not precise, although more modern developments do achieve more accuracy and higher quality compared to older ones, especially if they are adjusted to specialized fields or allow actions by a human being.
It should be noted that, when a corpus parallel to a pair of languages is available, it is possible to apply statistical techniques to an automatic translation system. In this way, it is possible to draw probabilities around which terms are most appropriate to replace others.
History of machine translation
The search for records and documentation to try to reconstruct and verify when and how automatic translation was born and how it has evolved takes us back several centuries.
As has been established, approximately in 1629 the French physicist, philosopher and mathematician René Descartes proposed the idea (and need) of instituting a universal language in order to promote communication without language barriers by appealing to guidelines of logic and symbols to allow that the same word or notion could be understood in multiple languages.
In 1954 , after increasing proposals to transcribe words into other languages, a demonstration was carried out that became popular in Spanish under the name of the Georgetown-IBM experiment . This commitment to automatic translation made it possible to translate more than six dozen phrases expressed in Russian into English and began a period of enthusiasm regarding investments to support research linked to automatic translation .
After a temporary drop in funding for the field, an increase in general interest in translation based on statistics was achieved in the 1980s .
It is constructive, in this framework, to read "Introduction to automatic translation" , a work launched by W. John Hutchins and Harold L. Somers , to deepen knowledge about the progress of methods and varieties in this regard.
Automatic translation favors increased productivity and, compared to human translation, is more economical, comfortable and faster.
Applications
In the contemporary world, machine translation achieves immense significance and considerable visibility.
A huge number of people, for example, take advantage of instant translation programs to communicate in real time with individuals of different nationalities with whom they do not share a language. This alternative, which is made available to Internet users for free, even serves to translate website content in a matter of seconds.
If we focus on specialized sectors, then automatic literary translation , the translation of medical texts and the translation of legal documents gain notoriety, to list some options for reference.
Both in the public sphere and in the private sphere there are translation tools and technologies, as can be seen when analyzing everyday reality. The resource of transferring material from one language to another is favorable in business environments to carry out wide-ranging communications that can be correctly interpreted, just as it is essential in the fields of scientific research and education . Nor should we lose sight of the fact that in the media and in the entertainment industry the exercise of translation is vital since it makes interviews and reports viable and guarantees that many viewers can watch, enjoy and understand foreign films.
Multilingual chatbots , automatic email translation and automatic dubbing services to produce multilingual content are further evidence of how translation has been put to good use in a multitude of ways in the era of the digital revolution . Of course, despite the improvements and growth, there is still much to investigate, do and specify after overcoming technical limitations and overcoming challenges related to culture, ethics, etc.
On certain occasions, machine translation presents limitations and has challenges to overcome in relation to the quality of the results it offers, given that they may be too literal, imprecise or erroneous.
Automatic translation classes
As machine translation systems expanded, diversified and modernized, numerous varieties emerged.
Among the most common types of machine translation systems are, first of all, those that develop a rule-based translation and the style of machine translation based on linguistic corpus . In this last group, the modalities of translation based on examples and statistical automatic translation coexist.
Thanks to technological advances, it has been possible to incorporate the so-called neural translation as a reliable option, a modality that invites us to appreciate the benefits of artificial intelligence (AI) . Platforms that provide agile voice-to-text translation services and useful applications for text-to-speech translation have also been consolidated.
If you want to explore and experiment with other possibilities, it is worth considering and comparing the advantages of asynchronous translation , simultaneous translation, cloud-based translation and minority language translation .