Skip to main content

Epic 3 - Admin login and Manual Audio Upload

To Be Done

Epic 3 implements admin login functionality and manual audio upload, allowing authenticated users to upload pre-recorded audio files instead of generating them via AI.

Objective​

Implement admin login authentication and add manual audio upload feature to allow authenticated users to upload their own pre-recorded audio files (MP3, WAV, etc.) and get a shareable URL.

User Flow:

Admin Login → Upload Audio File → Store in S3 → Generate Slug → Shareable URL

What You Get:

  • Upload pre-recorded audio files
  • Support for multiple audio formats (MP3, WAV, M4A)
  • Automatic slug generation
  • Shareable URL for uploaded audio
Building on Epic 1

Epic 3 complements the AI generation from Epic 1 by allowing manual uploads. Users can choose between AI generation or manual upload.

Priorities Covered​

  • Admin Login Authentication (New)
  • Manual Upload Feature (New)

Prerequisites from Epic 1:

  • AWS S3 storage configured (P3)
  • Slug generation working
  • User interface operational (P4)

What Changes​

Admin Login (Microsoft Identity)​

Authentication Required:

  • Admins must login with Microsoft Identity before accessing upload features
  • JWT token contains userId and userName
  • Token required for all upload API requests
  • createdBy/updatedBy fields auto-populated from token

Login Flow:

  • Admin navigates to upload page
  • Redirected to Microsoft Identity login if not authenticated
  • After successful login, JWT token stored
  • Token included in Authorization header for API requests

New Upload Option​

Homepage Updates:

  • Add "Upload Audio" button/tab alongside "Generate Audio"
  • File upload interface with drag-and-drop
  • Format validation (MP3, WAV, M4A, OGG)
  • File size limit (max 10MB per file)
  • Preview uploaded audio before confirming

Result Page​

Same Result Page as AI Generation:

  • Shows uploaded audio player
  • Displays shareable URL with slug
  • Copy link button
  • Download original file option

Technical Details​

Data Storage Updates​

Add new fields to generation records:

  • Upload Type: ai_generated or manual_upload
  • Original Filename: Original name of uploaded file
  • File Format: MP3, WAV, M4A, OGG, etc.
  • File Size: Size in bytes

Example Storage:

  • Upload Type: manual_upload
  • Original Filename: welcome-message.mp3
  • File Format: mp3
  • File Size: 2,456,789 bytes

API Flow​

Upload API Endpoint​

Endpoint: POST /api/v1/upload

Request (Multipart Form Data):

POST /api/v1/upload
Authorization: Bearer {access-token}
Content-Type: multipart/form-data

audioFile: [binary audio file]

Response:

{
  "success": true,
  "data": {
    "id": "550e8400-e29b-41d4-a716-446655440000",
    "slug": "welcome-message",
    "audioUrl": "https://s3.amazonaws.com/micdots/audio/550e8400.mp3",
    "playbackUrl": "https://micdots.com/play/welcome-message",
    "uploadType": "manual_upload",
    "originalFilename": "welcome-message.mp3",
    "fileFormat": "mp3",
    "fileSize": 2456789,
    "createdAt": "2024-11-02T12:00:00Z"
  }
}

Success Criteria​

Epic 3 is complete when:

  1. ✅ Admin login with Microsoft Identity works
  2. ✅ JWT token authentication is enforced for upload API
  3. ✅ createdBy/updatedBy auto-populated from JWT token
  4. ✅ Users can upload audio files
  5. ✅ Multiple formats are supported (MP3, WAV, M4A, OGG)
  6. ✅ File validation works (format, size)
  7. ✅ Audio is stored on AWS S3
  8. ✅ Unique slug is generated
  9. ✅ Shareable URL is created
  10. ✅ Audio playback works for uploaded files
  11. ✅ Upload type is tracked (ai_generated vs manual_upload)
  12. ✅ Manual testing completed and verified
  13. ✅ Deployed to production

Timeline​

Duration: 3 weeks

WeeksFocusDeliverables
Week 14Admin Login & AuthMicrosoft Identity integration, JWT auth
Week 15Upload API & ValidationFile upload endpoint, format validation
Week 16Testing & DeploymentValidation, testing, deployment

File Upload Specifications​

Supported Formats​

  • MP3: MPEG-1 Audio Layer 3 (.mp3)
  • WAV: Waveform Audio File Format (.wav)
  • M4A: MPEG-4 Audio (.m4a)
  • OGG: Ogg Vorbis (.ogg)

File Size Limits​

  • Maximum: 10MB per file
  • Recommended: Under 5MB for optimal performance
  • Duration: No limit, but recommend under 5 minutes

Validation Rules​

Format Validation:

  • Check file extension
  • Validate MIME type
  • Verify audio codec

Size Validation:

  • Reject files over 10MB
  • Show clear error message

Content Validation:

  • Ensure file is valid audio
  • Check for corruption
  • Verify playability

UI Mockup​

Homepage with Upload Option​

Two Options:

  1. Generate Audio (AI) - Text input → ElevenLabs generation
  2. Upload Audio (Manual) - File upload → Direct storage

Upload Interface:

  • Drag-and-drop zone
  • "Browse Files" button
  • Format requirements shown
  • File size limit displayed
  • Preview audio after selection
  • "Upload" button

Example:

┌─────────────────────────────────────────┐
│  [ Generate Audio ]  [ Upload Audio ]  │
├─────────────────────────────────────────┤
│                                         │
│   📁 Drag & drop your audio file here  │
│        or click to browse               │
│                                         │
│   Supported: MP3, WAV, M4A, OGG        │
│   Max size: 10MB                        │
│                                         │
│   [ Upload ]                            │
└─────────────────────────────────────────┘

Use Cases​

When to Use Manual Upload:

  • Pre-recorded professional voiceovers
  • Multilingual content not supported by AI
  • Custom music or sound effects
  • Client-provided audio content
  • Legacy audio files from old system
  • Audio edited externally (noise reduction, effects, etc.)

When to Use AI Generation:

  • Quick content creation
  • Consistent voice across content
  • No recording equipment available
  • Last-minute changes to text
  • Multiple language variations