File Import Guide
How to import CSV, SRT, and text files into AutoFlowCut
π Get Started in 5 Minutes Recommended
No complex setup needed. A simple text file is all you need to start!
Create a Text File
Write scene descriptions, one per line, in any text editor
The scholar walking toward the palace gate
A celebration scene with lanterns
Import the File
In AutoFlowCut, click "π Import" button β Select "Prompt Text File"
Generate Images!
Click "Start Generation" button at the bottom and Google Flow AI will automatically create images for each scene
π‘ Next: Start with a text file. When you need subtitles or consistent characters, use the CSV formats below.
π Supported File Types
One line per scene, simple format
Scene information with prompts, subtitles, and durations
Character/background/style definitions (for consistency)
Standard subtitle format, auto-converted to scenes
π Workflow Overview
See how the files connect at a glance
β Basic (Text only)
β Advanced (Character consistency)
scholar
scholar
π‘ Summary: When characters, scene_tag, style_tag in Scene CSV match name in Reference CSV, reference images are automatically applied.
π Import Modal Preview
This modal appears when you click the "π Import" button. Click on the file format you want.
Select a file format
Help buttons below each option:
π Prompt Text File Easiest
The simplest format - just write one scene description per line. Each line becomes a separate scene with a default duration.
Example
A young scholar reading under an ancient pine tree A merchant crossing a stone bridge at dawn Two warriors facing each other in a bamboo forest An old storyteller sitting by a campfire at night
π Scene CSV File
The Scene CSV file defines each scene's content, including prompts for image generation, subtitles, and timing.
π€ Why use CSV?
Text files are simple, but CSV gives you more control:
- β’ Subtitles separate from prompts (different content)
- β’ Character/Background/Style tags β Auto-match with reference images (see below)
- β’ Per-scene duration settings (3s, 5s, etc.)
π‘ When tags match references, you'll see β in the scene list
Required Columns
| Column | Description | Example |
|---|---|---|
| prompt | Image generation prompt | A scholar reading under a pine tree |
| subtitle | Subtitle text for the scene | Once upon a time, a young scholar lived in a village. |
| characters | Character names (semicolon-separated) | scholar;queen |
| scene_tag | Scene category tag | palace, village, journey |
| style_tag | Visual style tag | korean historical drama |
| duration | Scene duration in seconds | 5 |
CSV Example
prompt,subtitle,characters,scene_tag,style_tag,duration "A young scholar reading under an ancient pine tree, Joseon dynasty","Once upon a time, a young scholar lived in a village.",scholar,village,korean historical drama,5 "The scholar walking toward a grand palace gate at dawn","The scholar set off to the capital for the civil exam.",scholar,journey,korean historical drama,4 "Royal court filled with officials in formal attire","The palace in Hanyang was filled with dignified officials.",scholar;officials,palace,korean historical drama,5
| prompt | subtitle | characters | scene_tag | style_tag | duration |
|---|---|---|---|---|---|
| A young scholar reading under an ancient pine tree | Once upon a time, a young scholar lived in a village. | scholar | village | korean historical drama | 5 |
| The scholar walking toward a grand palace gate at dawn | The scholar set off to the capital for the civil exam. | scholar | journey | korean historical drama | 4 |
| Royal court filled with officials in formal attire | The palace in Hanyang was filled with dignified officials. | scholar;officials | palace | korean historical drama | 5 |
πΌοΈ Reference CSV File
The Reference CSV file defines characters, backgrounds, and styles that can be used across scenes for consistent visual generation.
β No images required!
You can leave image_path empty in the CSV. Add images later using:
- β’ Direct upload - Click image area in References tab β Select file
- β’ AI generation - Generate reference images with Flow AI based on description
- β’ Add later - Set up the structure now, add images anytime
π Tag Matching System (Very Important!)
The name field in Reference CSV automatically matches with tags in Scene CSV.
When generating images, matched reference images are uploaded to Flow AI to maintain consistent style/characters.
type: charactername: scholar
characters column
type: scenename: palace
scene_tag column
type: stylename: korean historical drama
style_tag column
Required Columns
| Column | Description | Example |
|---|---|---|
| type | Reference type (character / scene / style) | character |
| name | Name for tag matching (must match Scene CSV) | scholar, palace, korean historical drama |
| image_path | Path or URL to reference image (optional) | ./images/scholar.png |
| description | Detailed description (for image generation) | Young scholar in traditional dopo robe |
CSV Example
type,name,image_path,description character,scholar,,"A young Korean scholar in white dopo robe, topknot with gat hat, gentle expression" character,queen,,"Joseon dynasty queen in elaborate royal hanbok with phoenix embroidery" scene,palace,,"Grand Joseon palace interior with wooden pillars and dancheong paintwork" scene,village,,"Traditional Korean village with thatched-roof houses and mountain backdrop" style,korean historical drama,,"Korean historical drama style, cinematic composition, warm color palette"
| type | name | image_path | description |
|---|---|---|---|
| character | scholar | A young Korean scholar in white dopo robe, topknot with gat hat | |
| character | queen | Joseon dynasty queen in elaborate royal hanbok | |
| scene | palace | Grand Joseon palace interior with wooden pillars | |
| style | korean historical drama | Korean historical drama style, cinematic composition |
π¬ Subtitle SRT File
Import standard SRT subtitle files. Each subtitle block is automatically converted to a scene with the subtitle text used as both the prompt and subtitle. Duration is calculated from the timecodes.
SRT Structure
1 00:00:00,000 --> 00:00:05,000 Once upon a time, a young scholar lived in a village. 2 00:00:05,000 --> 00:00:09,000 Every day he studied, reading books under the pine tree. 3 00:00:09,000 --> 00:00:14,000 One day, he decided to travel to Hanyang for the civil exam.
Each numbered block becomes one scene. The timecode determines scene duration automatically.
ποΈ Auto-generate SRT from TTS Services
These TTS (Text-to-Speech) services automatically generate SRT subtitle files along with the audio. Get your narration and subtitles ready to import directly into AutoFlowCut.
π‘ Tip: After generating audio with a TTS service, download the SRT file and import it into AutoFlowCut for perfectly timed scenes.
π€ Auto-generate CSV with AI
Paste the prompts below into Claude, ChatGPT, Gemini, or other AI tools along with your story to automatically generate CSV files.
π Scene CSV Generation Prompt
You are a scene breakdown assistant for AutoFlowCut video production. Given a story or script, create a Scene CSV file with these columns: - prompt: English scene description for AI image generation (include composition, lighting, mood, camera angle) - subtitle: Subtitle text for the scene (concise, under 50 chars) - characters: Character names in the scene (semicolon-separated if multiple, e.g., scholar;queen) - scene_tag: Background/location tag (use consistent tags like palace, forest, village) - style_tag: Art style (keep consistent, e.g., korean historical drama, ghibli, cinematic) - duration: Scene duration in seconds (3-5 for normal, 2 for quick cuts, 6+ for dramatic) Rules: 1. Each row = one visual scene (single image frame) 2. Keep prompts descriptive but under 200 characters 3. Use consistent character names throughout (these match reference images) 4. Group similar locations under one scene_tag 5. Consider pacing: vary duration for dramatic effect 6. Output as CSV with header row, quote values containing commas Example output: prompt,subtitle,characters,scene_tag,style_tag,duration "A young scholar reading under an ancient pine tree, Joseon dynasty","Once upon a time, a young scholar lived in a village.",scholar,village,korean historical drama,5 Now break down this story into scenes:
πΌοΈ Reference CSV Generation Prompt
You are a reference planning assistant for AutoFlowCut. Given a story, create a Reference CSV with these columns: - type: Reference type (character, scene, or style) - name: Name for tag matching (MUST match scene CSV tags exactly) - image_path: Leave empty (images will be generated by Flow AI) - description: Detailed description for generating the reference image For characters: List all unique characters with their visual appearance For scenes: List all unique locations/settings mentioned For styles: Suggest 1-2 art styles that fit the story mood IMPORTANT: The 'name' field must match exactly with: - character type β matches 'characters' column in scene CSV - scene type β matches 'scene_tag' column in scene CSV - style type β matches 'style_tag' column in scene CSV Example output: type,name,image_path,description character,scholar,,"Young Korean scholar in white dopo robe with topknot and gat hat" scene,palace,,"Grand Joseon palace throne room with wooden pillars and dancheong" style,korean historical drama,,"Korean historical drama style, cinematic composition, warm tones" Now create the reference list for this story:
π‘ AI Tool Tips
Great for long stories. Use Artifacts feature to download CSV directly.
Use Code Interpreter to create and download CSV. Preview as table.
Works with Google Docs. Easy export to spreadsheets.
π How to Import Files
Open AutoFlowCut desktop app
Launch AutoFlowCut and sign in with your Google account
Click the Import button
Located in the sidebar or toolbar
Select your file
Choose a CSV, SRT, or TXT file from your computer
Review and generate
Check the imported scenes and start generating images with Google Flow AI
π₯ Download Sample Files
Download these sample files to get started quickly: