Skip to content

API.md

Latest commit

 

History

History
243 lines (187 loc) · 4.17 KB

API.md

File metadata and controls

243 lines (187 loc) · 4.17 KB
Raw

Video2Notes API Documentation

Overview

Video2Notes provides a RESTful API organized via Flask blueprints for managing video processing workflows, file operations, and interactive components.

Base URL

http://localhost:5100

API Endpoints

Core Workflow Routes

GET /

Main application page with video upload and workflow configuration interface.

POST /workflow/start

Start a new video processing workflow.

Request Body:

{
  "video_path": "/path/to/video.mp4",
  "extract_audio": true,
  "skip_roi": false,
  "roi_timestamp": 300,
  "do_label_speakers": true,
  "do_refine_notes": true,
  "refine_notes_llm": "openai/gpt-4o-2024-08-06",
  "do_split": false,
  "timestamp_file": null
}

Response:

{
  "success": true
}

GET /workflow/progress

Workflow progress page with real-time updates via Server-Sent Events.

GET /workflow/status

Get current workflow status as JSON.

Response:

{
  "status": "running",
  "progress": 45,
  "current_step": "Transcribing audio",
  "logs": ["🚀 Workflow started", "📹 Processing video..."],
  "interactive_stage": null,
  "available_files": []
}

POST /workflow/stop

Stop the current running workflow.

Response:

{
  "success": true
}

File Management Routes

POST /files/upload

Upload a video file for processing.

Request:

  • Method: POST
  • Content-Type: multipart/form-data
  • Body: Video file

Response:

{
  "success": true,
  "filename": "uploaded_video.mp4",
  "path": "/path/to/uploaded_video.mp4"
}

GET /files/browse

Browse available files on the server.

Query Parameters:

  • path - Directory path to browse (optional)

Response:

{
  "files": [
    {
      "name": "video1.mp4",
      "path": "/videos/video1.mp4",
      "size": 1048576,
      "type": "file"
    }
  ],
  "current_path": "/videos"
}

GET /download/

Download generated files.

Parameters:

  • filename - Name of the file to download

Response:

  • Binary file download

SharePoint Integration Routes

POST /files/sharepoint/get-download-links

Get download links for a SharePoint video.

Request Body:

{
  "sharepoint_video_url": "https://sharepoint.com/video.mp4"
}

POST /files/sharepoint/download

Download a video from SharePoint.

Request Body:

{
  "direct_download_link": "https://sharepoint.com/direct/video.mp4",
  "output_filename": "video.mp4"
}

GET /files/sharepoint/download-progress

Get SharePoint download progress via SSE.

Interactive Component Routes

Slide Selection Routes (/slides/)

GET /slides/selector

Open the interactive slide selection interface.

POST /slides/save

Save selected slides.

Request Body:

{
  "selected_slides": [0, 2, 5, 8]
}
POST /slides/update-vocabulary

Enable/disable vocabulary extraction from slides.

Request Body:

{
  "extract_vocabulary": true
}
GET /slides/download-vocabulary

Download extracted vocabulary file.

Speaker Labeling Routes (/speakers/)

GET /speakers/labeler

Open the interactive speaker labeling interface.

POST /speakers/get-speakers

Get list of speakers from the transcript.

POST /speakers/get-audio-segment

Get audio segment for a specific utterance.

Request Body:

{
  "start": 10.5,
  "end": 15.3
}
POST /speakers/label-speaker

Label a speaker with a name.

Request Body:

{
  "speaker_id": "SPEAKER_00",
  "speaker_name": "John Doe"
}
POST /speakers/save-labeled-notes

Save the notes with labeled speaker names.

Server-Sent Events (SSE)

GET /workflow/stream

Real-time workflow progress updates.

Event Format:

data: {"progress": 50, "message": "Extracting slides...", "status": "running"}

GET /files/sharepoint/download-progress

Real-time SharePoint download progress.

Event Format:

data: {"progress": 75, "message": "Downloading: 75%"}

Response Codes

  • 200 - Success
  • 400 - Bad Request (invalid parameters)
  • 404 - Not Found
  • 500 - Internal Server Error

Error Response Format

{
  "success": false,
  "error": "Error message description"
}