Scribe v2 Realtime

# 1. Request a signed upload URL
curl https://opentranscription.io/api/v1/uploads \
  -H "Authorization: Bearer $OT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "file_name": "interview.mp3", "file_size": 3456789, "mime_type": "audio/mpeg" }'
# → { "upload_url": "...", "file_path": "...", "expires_at": "..." }

# 2. Upload the audio bytes to the signed URL
curl -X PUT "$UPLOAD_URL" --upload-file interview.mp3

# 3. Create the transcription job
curl https://opentranscription.io/api/v1/transcriptions \
  -H "Authorization: Bearer $OT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "file_path": "$FILE_PATH",
    "model": "elevenlabs/scribe-v2-realtime",
    "language": "af",
    "webhook_url": "https://your-app.com/webhooks/ot"
  }'
# → we POST a signed event to your webhook_url when the job finishes — no polling required

{
  "id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
  "status": "completed",
  "transcript": {
    "text": "Welcome to the meeting, everyone.",
    "language": "en",
    "confidence": 0.97,
    "segments": [
      {
        "id": 0,
        "start": 0,
        "end": 3.5,
        "text": "Welcome to the meeting, everyone."
      }
    ],
    "words": [
      {
        "word": "Welcome",
        "start": 0,
        "end": 0.42,
        "confidence": 0.99
      }
    ]
  }
}

# We POST a signed event to your webhook_url when a job finishes.
# Headers:
#   X-OT-Signature: t=1736937042,v1=5257a869e7eceb6c6c2c2a1b4d6e9f8c2a1b4d6e9f8c2a1b4d6e9f8c2a1b4d6e
#   X-OT-Event-Id: evt_9f8c2a1b4d6e
#   X-OT-Event-Type: transcription.completed
#
# Body (thin event — fetch the transcript with GET /transcriptions/{data.id}):
{
  "id": "evt_9f8c2a1b4d6e",
  "event": "transcription.completed",
  "created": "2025-01-15T10:30:42.000Z",
  "data": {
    "id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "status": "completed",
    "error_code": null
  }
}
#
# Verify: HMAC-SHA256 over "<t>.<raw_body>" with your signing secret,
# constant-time compare against any v1, and reject if |now - t| > 300s.