Translate and review Japanese product listings with OpenAI GPT-5-mini

Quick Overview

This workflow collects a Japanese product listing via an n8n form, translates it into English with OpenAI, then runs a second OpenAI pass to review and correct the translation against the original before displaying the reviewed listing and any fixes.

How it works

1. Receives a product name, Japanese description, and target market when an n8n form is submitted.
2. Adds your store name and a fixed translation glossary to the submitted data.
3. Uses OpenAI (chat model) to translate the listing into a structured English title, description, and bullet points while preserving facts and applying the glossary.
4. Uses OpenAI again to compare the English draft with the Japanese original, correct issues (missing facts, altered numbers, added claims, awkward English), and output a quality score plus a list of fixes.
5. Displays the reviewed English listing, quality score, and the reviewer’s issues list on a completion page.

Setup

1. Add an OpenAI API credential and select the model in the OpenAI Chat Model node used by both translation and review steps.
2. Update the store name and glossary mappings in the Set Store Details step to match your brand terms and preferred translations.
3. Open and use the generated n8n form URL, then submit a Japanese listing to test the reviewed output.