🚀 Project Setup Guide

🌐 Live Demo

👉 Click here to view the live site

🛠️ Tech Stack

📦 Dependency Setup

A. Automatic Setup

chmod +x setup.sh
./setup.sh

B. Manual Setup

1. Make sure Node.js is installed.
2. In the project root, run:

   npm install

3. Copy .env.example to .env.local and fill in your own API keys.

---

🔑 Environment Variables

Refer to .env.example and fill in your credentials in .env.local.

• 🟢 Supabase: supabase.com
• 🤖 OpenAI: openai.com
• 🗣️ Azure AI Speech: azure.microsoft.com
• 📰 Alpha Vantage: alphavantage.co

---

🗄️ Database Setup

1. Get familiar with Supabase.
2. Enable the pgvector extension:
   Supabase vector columns guide
3. Run the following script in the Supabase SQL editor:

-- Enable pgvector if not already enabled
CREATE EXTENSION IF NOT EXISTS vector;
-- Drop old function if exists
DROP FUNCTION IF EXISTS match_documents(vector, integer, jsonb);
-- Create documents table
CREATE TABLE IF NOT EXISTS documents (
  id uuid PRIMARY KEY DEFAULT gen_random_uuid(),
  content text,
  embedding vector(1536),
  created_at timestamp DEFAULT now(),
  metadata jsonb
);
-- Optional: create index for jsonb metadata tag search
CREATE INDEX IF NOT EXISTS idx_metadata_tags ON documents USING gin ((metadata->'tags'));
-- Create match_documents function
CREATE OR REPLACE FUNCTION match_documents (
  query_embedding vector(1536),
  match_count int,
  filter jsonb DEFAULT '{}'::jsonb
)
RETURNS TABLE (
  id uuid,
  title text,
  content text,
  metadata jsonb,
  similarity float
)
LANGUAGE plpgsql
AS $$
BEGIN
  RETURN QUERY
  SELECT
    docs.id,
    docs.metadata->>'title' AS title,
    docs.content,
    docs.metadata,
    1 - (docs.embedding <=> query_embedding) AS similarity
  FROM documents AS docs
  WHERE
    (
      ((filter - 'tags') = '{}'::jsonb) OR (docs.metadata @> (filter - 'tags'))
    )
    AND
    (
      NOT (filter ? 'tags') OR 
      (
        (docs.metadata ? 'tags') AND
        (jsonb_typeof(docs.metadata->'tags') = 'array') AND
        (docs.metadata->'tags' ?| ARRAY(SELECT jsonb_array_elements_text(filter->'tags')))
      )
    )
  ORDER BY docs.embedding <=> query_embedding
  LIMIT match_count;
END;
$$;

---

📥 Data Retrieval & Indexing

• Configure the news topics you want to fetch in script/process-indexing.ts.
• For valid parameters, see: Alpha Vantage News Sentiment docs
• Podcast parameters (e.g., account, topic) are preset and cannot be changed.
• To fetch and index data, run:

  npm run test ./script/process-indexing.ts

---

🖥️ Run the Server

npm run dev

---

🗃️ Database and Types

• Check types/supabase.ts for the current schema types.
• After changing the database schema, regenerate types with:

  npm run supabase:generate-types

---

🏗️ System Structure