You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
feat: add German language support (de-DE) following new i18n system
Refactored the initial German language support to align with the newly
introduced i18n system.
Changes:
- Added de-DE.json with all translation keys.
- Registered de-DE in supportedLocales.
- Localized fallback strings in the generation pipeline.
- Updated UI components to dynamically use supportedLocales.
- Added Azure TTS German voices.
- Made prompts generic for multiple languages.
Closes#215
Copy file name to clipboardExpand all lines: lib/generation/prompts/templates/requirements-to-outlines/system.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -94,7 +94,7 @@ When a slide scene needs an image or video but no suitable PDF image exists, mar
94
94
-**Image IDs**: use `"gen_img_1"`, `"gen_img_2"`, etc. — IDs are **globally unique across the entire course**, NOT reset per scene
95
95
-**Video IDs**: use `"gen_vid_1"`, `"gen_vid_2"`, etc. — same global numbering rule
96
96
- The prompt should describe the desired media clearly and specifically
97
-
-**Language in images**: If the image contains text, labels, or annotations, the prompt MUST explicitly specify that all text in the image should be in the course language (e.g., "all labels in Chinese" for zh-CN courses, "all labels in English" for en-US courses). For purely visual images without text, language does not matter.
97
+
-**Language in images**: If the image contains text, labels, or annotations, the prompt MUST explicitly specify that all text in the image should be in the course language (e.g., "all labels in German" for de-DE courses). For purely visual images without text, language does not matter.
98
98
- Only request media generation when it genuinely enhances the content — not every slide needs an image or video
99
99
- Video generation is slow (1-2 minutes each), so only request videos when motion genuinely enhances understanding
100
100
- If a suitable PDF image exists, prefer using `suggestedImageIds` instead
@@ -280,7 +280,7 @@ You must output a JSON array where each element is a scene outline object:
280
280
"projectDescription": "Brief description of what students will build/accomplish",
text='Beginnen wir mit einer projektbasierten Lernaktivität. Wählen Sie Ihre Rolle, prüfen Sie das Aufgabenboard und arbeiten Sie zusammen, um das Projekt abzuschließen.';
text='Erforschen wir nun dieses Konzept durch eine interaktive Visualisierung. Versuchen Sie, mit den Elementen auf der Seite zu interagieren, um zu sehen, wie sie sich verändern.';
0 commit comments