Backend API for the Funnel audio note summarizer app built with Deno.

• Deepgram Integration Guide - How we use Deepgram for real-time transcription
• Deepgram Sequence Diagram - Visual flow of the transcription process
• Recording Flow - Overall recording and processing flow

1. Copy .env.example to .env and add your API keys:

   cp .env.example .env

2. Add your API keys to .env:

   • OPENAI_API_KEY - For Whisper audio transcription
   • ANTHROPIC_API_KEY - For Claude text summarization

Run the development server with auto-reload:

deno task dev

Transcribes audio files using OpenAI Whisper.

Request:

• Method: POST
• Content-Type: multipart/form-data
• Body: audio field with audio file (mp3, mp4, wav, m4a)
• Max file size: 25MB

Response:

{ "transcript": "The transcribed text...", "duration": 123.45 }

Generates bullet point summary using Anthropic Claude.

Request:

• Method: POST
• Content-Type: application/json
• Body:

{ "transcript": "The text to summarize..." }

Response:

{ "bulletSummary": [ "First key point", "Second key point", "Action item or task" ] }

1. Create a new project on Deno Deploy
2. Set environment variables in the project settings:
   • OPENAI_API_KEY
   • ANTHROPIC_API_KEY
3. Deploy using GitHub integration or Deno Deploy CLI

Run in production:

deno task start

• deno task dev - Run development server with watch mode
• deno task start - Run production server
• deno task fmt - Format code
• deno task lint - Lint code
• deno task check - Type check
• deno task precommit - Run all checks before committing