Welcome to DocuTranslate! Follow these steps to translate your documents:
Select Workflow
First, choose the translation process you need from the top of the settings panel. Different workflows are for different file types:
Markdown-based: For translating PDF, markdown, images, etc.
Plain Text: For translating .txt and other plain text files.
JSON: For translating specific fields in .json files.
DOCX: For translating .docx files.
XLSX: For translating .xlsx spreadsheet files.
SRT Subtitle: For translating .srt subtitle files.
EPUB: For translating .epub ebook files.
New Feature: \"Auto-select workflow\" is on by default. Just upload your file, and the system will automatically pick the right workflow for you.
Configure Parameters
Based on your chosen workflow, complete the necessary settings. All settings are automatically saved in your browser.
Parsing Configuration (Only for \"Markdown-based\" workflow):
Parsing Engine: Choose an engine to convert your file (like a PDF) into a translation-friendly Markdown format. Not needed if your file is already Markdown.
Mineru Token: If you select the minerU engine, you need to enter your token here.
DOCX/XLSX/SRT/EPUB Options (For their respective workflows):
Insert Mode: Define how the translation is placed in the document. You can \"Replace\" the original, \"Append\" after it, or \"Prepend\" before it.
Separator: When using \"Append\" or \"Prepend\", this is used to separate the original and translated text.
JSON Path Configuration (Only for \"JSON\" workflow):
JSON paths to translate: Enter one JSONPath expression per line to specify which fields to translate.
E.g., $..description translates all values with the key 'description'. $.items[0].name translates the name of the first item.
Translation Model:
Platform/API Base URL/API Key/Model ID: Configure the AI translation service you want to use.
Refer to the platform's documentation for Model IDs. It's recommended to use non-inference models or mixed-inference models (with thinking turned off).
Translation Configuration:
Target Language/Custom Prompt: Specify the target language and any additional instructions.
Thinking Mode: Sets whether a mixed-inference model should 'think'. Supported for Zhipu's glm4.5 series, Alibaba's qwen3 series, VolcEngine's seed1.6 series, etc. Disabling is recommended.
Advanced Parameters:
Chunk Size/Concurrency/Temperature: The size of text chunks sent to the AI, number of concurrent requests, and creativity level. Default values usually work fine.
Upload File
In the task list on the right, click or drag your document into the file drop area.
Start Translation
Once the file is selected, click the Start Translation button on the task card. The system will start processing, and you can see real-time progress in the log area.
Review & Download
After translation is complete, action buttons will appear on the task card:
Preview: Opens a side panel for a side-by-side preview of the original and translation (for reference only).
Download: Download the translation in various formats, including PDF, DOCX, XLSX, HTML, and Markdown.
Tip: All your settings are automatically saved in your browser's local storage for your next visit.
",
+ "tutorialModalBody": "
Video tutorials can be found by searching docutranslate on Bilibili.
Welcome to DocuTranslate! Follow these steps to translate your documents:
Select Workflow
First, choose the translation process you need from the top of the settings panel. Different workflows are for different file types:
Markdown-based: For translating PDF, markdown, images, etc.
Plain Text: For translating .txt and other plain text files.
JSON: For translating specific fields in .json files.
DOCX: For translating .docx files.
XLSX: For translating .xlsx spreadsheet files.
SRT Subtitle: For translating .srt subtitle files.
EPUB: For translating .epub ebook files.
HTML: For translating .html files.
New Feature: \"Auto-select workflow\" is on by default. Just upload your file, and the system will automatically pick the right workflow for you, simplifying the process.
Configure Parameters
Based on your chosen workflow, complete the necessary settings. All settings are automatically saved in your browser.
Parsing Configuration (Only for \"Markdown-based\" workflow):
Parsing Engine: Choose an engine to convert your file (like a PDF) into a translation-friendly Markdown format. Not needed if your file is already Markdown.
Mineru Token: If you select the minerU engine, you need to enter your token here.
DOCX/XLSX/SRT/EPUB/HTML Options (For their respective workflows):
Insert Mode: Define how the translation is placed in the document. You can \"Replace\" the original, \"Append\" after it, or \"Prepend\" before it.
Separator: When using \"Append\" or \"Prepend\", this is used to separate the original and translated text.
JSON Path Configuration (Only for \"JSON\" workflow):
JSON paths to translate: Enter one JSONPath expression per line to specify which fields to translate.
E.g., $..description translates all values with the key 'description'. $.items[0].name translates the name of the first item.
Translation Model:
Platform/API Base URL/API Key/Model ID: Configure the AI translation service you want to use.
Refer to the platform's documentation for Model IDs. It's recommended to use non-inference models or mixed-inference models (with thinking turned off).
Translation Configuration:
Target Language/Custom Prompt: Specify the target language and any additional instructions.
Thinking Mode: Sets whether a mixed-inference model should 'think'. Supported for Zhipu's glm4.5 series, Alibaba's qwen3 series, VolcEngine's seed1.6 series, etc. Disabling is recommended.
Advanced Parameters:
Chunk Size/Concurrency/Temperature: The size of text chunks sent to the AI, number of concurrent requests, and creativity level. Default values usually work fine.
Upload File
In the task list on the right, click or drag your document into the file drop area.
Start Translation
Once the file is selected, click the Start Translation button on the task card. The system will start processing, and you can see real-time progress in the log area.
Review & Download
After translation is complete, action buttons will appear on the task card:
Preview: Opens a side panel for a side-by-side preview of the original and translation (for reference only).
Download: Download the translation in various formats, including PDF, DOCX, XLSX, HTML, and Markdown.
Tip: All your settings are automatically saved in your browser's local storage for your next visit.
",
"tutorialUnderstandBtn": "Got it",
"contributorsModalTitle": "Thanks for Contributing",
"contributorsPara1": "DocuTranslate is an open-source project! The community's needs and usage are what drive its progress.",
@@ -264,7 +289,8 @@
"pdf_preparing": "Preparing PDF, please wait...",
"pdf_print_failed": "Automatic printing failed. Please print manually from the preview.",
"pdf_fetch_failed": "Failed to fetch HTML content, cannot generate PDF.",
- "init_failed_alert": "Page initialization failed. Please check if the backend service is running and refresh the page.",
+ "init_i18n_failed_alert": "Failed to load interface translations. Please check your network connection or contact an administrator.",
+ "init_failed_alert": "Initialization failed, could not connect to the backend service. Please ensure the service is running and refresh the page.",
"admin_tasklist_failed": "Could not load task list from server. Please check backend connection.",
"btn_startTranslation": "Start Translation",
"btn_initializing": "Initializing...",
diff --git a/docutranslate/static/index.html b/docutranslate/static/index.html
index f046344..708d303 100644
--- a/docutranslate/static/index.html
+++ b/docutranslate/static/index.html
@@ -1 +1 @@
-
DocuTranslate - 交互式文档翻译