The correction pass

Machine translation post-editing software — correct the output, or authorize the draft.

Machine translation post-editing software is where a person corrects machine output before delivery. Full post-editing, the work ISO 18587 describes, aims at a human-quality text; light post-editing only makes the text usable. Aquilla does not run that pass: a person authorizes the draft in the workspace, and Aquilla does not connect an engine hub or a marketplace of post-editors.

The direct answer

What is machine translation post-editing software?

It is the editor for correcting machine translation, split into a light pass and a full pass. ISO 18587:2017 specifies the full pass, and Smartling, XTM Cloud, and Smartcat sell the workspace around it, often with a connected engine. Aquilla does not: agents propose a draft, and a person authorizes meaning before it is treated as finished.

People searching for machine translation post-editing software are choosing a correction product. They want a service level, an engine, and sometimes a post-editor they can hire. A workspace that reviews its own draft has not become that product.

The expert-in-the-loop page answers a different question: where a qualified person should sit so the rest of the work can follow. Use that page for the control point. Use this page for the post-editing tool.

What the search is actually choosing

Post-editing software is a job list — not “a person looked at the draft.”

Searchers mix a light pass, a full ISO 18587 pass, an engine connection, and a marketplace. Aquilla covers the authorization of a draft. It does not cover the rest.

JobAquillaTypical result for this search
Light post-editing, enough to use the textNo — Aquilla does not sell a light post-editing level.Yes — a faster pass when the text only needs to be usable
Full post-editing to the quality of a human translationNo — Aquilla is not an ISO 18587 post-editing process.Yes — ISO 18587 describes that full human pass
Pre-translate from DeepL, Google, Microsoft, or another engineNo — Aquilla does not connect an MT engine hub.Yes — Smartling, XTM, and Smartcat connect engines
Hire a post-editor from a marketplaceNo — the reviewers are the people you already trust.Yes — Smartcat’s marketplace is built for that
Score the edit with edit distance or a post-editing effort numberNo — Aquilla does not report post-editing effort.Often a quality-estimation or effort metric on the file
A person authorizes the draft before it is treated as finishedYes — experts authorize meaning in the workspace.Yes — a post-editor corrects the machine output

Keep the post-editing tool when the pass is the job

Full post-editing and light post-editing are different jobs

Light post-editing makes machine output usable. Full post-editing, the work ISO 18587 specifies, aims at a text that could have been translated by a person. Machine translation post-editing software is the editor for that correction. The standard describes the process and the post-editor’s competence. It is not a program you install.

What stays in Smartling, XTM, or Smartcat

Those platforms pre-translate from connected engines, then put a linguist on the output. Some add a glossary, a quality check, or a marketplace of post-editors. Keep that stack when the contract is light versus full post-editing, an engine connection, or a hired post-editor. Aquilla does not replace it.

What Aquilla covers

Agents propose a draft from the project’s context. A person authorizes meaning before the draft is treated as finished. Living Memory can apply a decision the team has already approved to a later passage. That is review inside the workspace, not a scored pass over a machine-translation file.

Related, and not this page

Expert-in-the-loop translation is where the expert sits in the work, early enough to steer it. The AI-assisted translation workflow page is the sequence from context to export. This page is only the post-editing product: the editor, the service level, and the engine.

Post-editing vs the expert’s seat

This page is the correction product. The expert page is a different control point.

Machine translation post-editing starts from a batch of machine output and asks a linguist to repair it. Expert-in-the-loop translation asks the expert to steer decisions early enough that agents can apply them. Aquilla is built for the second job. It will not pretend the first job is included.

If the contract names light or full post-editing, an engine, or a hired post-editor, keep Smartling, XTM, Smartcat, or the editor your client already accepts.

Common questions

Machine translation post-editing FAQ

What is machine translation post-editing software?

It is the program where a person corrects machine-translation output before delivery. The usual split is light post-editing, which makes the text usable, and full post-editing, which aims at the quality of a human translation. Smartling, XTM Cloud, and Smartcat are platforms that pre-translate and then hand the output to a linguist. Aquilla does not sell that correction pass.

What is the difference between light and full post-editing?

Light post-editing fixes enough that someone can use the text. Full post-editing goes further: the result should read as a human translation would. ISO 18587:2017 specifies the full pass and the competence of the post-editor. Aquilla does not offer either level as a service, and it does not certify post-editors.

Does ISO 18587 require a specific program?

No. ISO 18587 is a process standard for full human post-editing of machine translation. It does not name a CAT tool. A team can follow it in whichever editor holds the bilingual file. Aquilla is not certified to that standard, and using Aquilla is not a claim that a project meets it.

Does Aquilla replace Smartling, XTM, or Smartcat for post-editing?

No. Those products connect machine-translation engines and give a linguist the output to correct, sometimes with a marketplace of reviewers. Aquilla does not connect DeepL, Google Translate, Microsoft Translator, or a custom engine hub, and it does not hire the reviewer. The people who authorize a draft are the people already on the project.

Can I connect an MT engine and post-edit the file in Aquilla?

No. Aquilla does not import a machine-translation file for a post-editing pass, and it does not connect an outside engine so you can correct that engine’s output segment by segment. Agents propose a draft inside the workspace. A person authorizes it there.

Is expert-in-the-loop the same as post-editing?

No. Post-editing asks a linguist to repair a batch of machine output. Expert-in-the-loop, on its own page, is a different control point: expert decisions steer the work early enough to apply elsewhere, instead of cleaning a finished batch. This page is the post-editing search. That page is the control-point search.

Does Aquilla measure post-editing effort?

No. Aquilla does not report edit distance, a post-editing effort score, or a light-versus-full recommendation. If a delivery needs that number, measure it in the tool that already produces it.

How much does this cost in Aquilla?

Anyone can start on the Free plan with a weekly allowance of AI assistance. Bible-translation teams and ETEN affiliates get team-plan access at no cost — contact us to activate your organization’s workspace.

Related topics

Authorize the draft. Keep the post-editing tool you already trust.

Open a Free plan and review a draft in the browser. Bible-translation teams and ETEN affiliates can contact us for team-plan access at no cost. Aquilla will not replace a post-editing bench.