Augmented translation and authorship’s role on this context

Augmented Translation: What's new in this term? When we talk about Augmented Translation, we're simply referring to the use of technologies, such as Artificial Intelligence, in the translation process.

Objectively, it involves a set of actions and activities facilitated by optimized platforms, like the Bureau Works, while the final decisions always rest with the human translator.

The translation industry is undergoing a transformation, with augmented reality at the forefront. It's reshaping how linguists work, encouraging them to enhance their approach with technology for more specialized, fast, and consistent translations.

Opportunities provided by Augmented Translation‍

What can Augmented Translation offer to translators or translation agencies? It provides a richer translation experience closely tied to the use of LLMs. The opportunities arise from the effective interaction between the translator and these models.

• Quality and Results in Terms of Business ‍

Augmented Translation enhances quality, production levels, and overall translation process agility.

Productivity and greater efficiency result from eliminating outdated processes in translation. For example, no more checking numerous sources for a basic word comparison or spending hours meticulously searching through manual texts.

Quality is felt by both the translator in their workflow and the clients. Client quality translates to business performance results achieved through the ranking benefits that LLMs offer. For the human translator, quality comes from an enriched translation experience, identifying errors and improving precision, which is primarily promoted by model traceability.

• Semantic Improvement

A prominent opportunity afforded by augmentation is what we call "translation smells". Basically a semantic checker. It looks for meaning problems within the text and signals them to the translator.

For instance, comparing the sentences:

• Lucya has gone to her home country.
• Lucya went to her birth country.
  

At first glance, the sentences have the same meaning. But if you have "home country" in your glossary to refer to the country of origin, it suggests a match by the platform instead of "birth country," "native country," or another option.

Semantic gain is connected to AI's ability to offer intelligent suggestions, providing more compatible alternatives for words or phrases within your project's context. The better the tool is trained, the more adept it becomes at suggesting alternatives. And more faithful to the translator preferences.

Beyond a semantic error corrector, translation smells also have the sensitivity to redflags aspects like gender bias, omissions, additions, spellings, missing tags, etc.

In summary, augmented translation allows the translator to understand if they are using the right tone, for which audience, and with the correct syntax.

• Proofreading‍

The proofreading capacity of augmented translations is another significant opportunity for the translator. It becomes much more efficient to identify typos and spelling errors that may escape the human eye. In this aspect, machine intelligence is more reliable. Of course, LLMs provide a database that surpasses the storage capacity of a human mind.

All these opportunities are also related to adding more style and authorship to the content. Despite all the suggestions, the translator holds the power to accept or reject alternatives, having the final say. Massively automating processes can be liberating.

Authorship’s place in this context

Within this context of augmented translation, the question of authorship generates significant debate among translators. "Am I still the author or just a validator, a kind of post-editor of the text?"

Human translators have the irreplaceable ability to interpret and assimilate culture. This should be valued and emphasized whenever we talk about the interaction between technology and translation.

The intention is never to replace the human translator but to enhance their tools and provide a comprehensive and useful view for each translation project's context. How?

Through a vast set of data, the translator can understand various aspects: "How was this term translated before?" "Has this sentence been translated in a similar thematic context?" "How often has this type of content been translated before?" "What is the traceability of this word?"

This set of data achieves a traceability that the human translator cannot compete with. And it's not productive for them to do so! That's why we always emphasize: integration, AI training, and alliance with human skills will define the future of translation.

The translator's role is to be a gatekeeper, a term widely used in communication to define someone who knows the criteria and filters or edits according to these criteria. This is the place of authorship in this context.

Augmented Translation frees up mental space for the translator to think about much more sensitive and important details of the translation. And it's in this sensitivity and discernment that the translator continues to affirm their place and invaluable value in the translation industry.

‍