prototype-ai

Setup virtual environment

python -m venv env

Activate virtual env

source env/bin/activate

Install dependencies

pip install -r requirements.txt

Setup Environment variables

OPEN_AI_PROJECT_ID = ""
OPEN_AI_ORGANIZATION_ID = ""
OPEN_AI_API_KEY = ""

GLADIA_API_KEY = ""

POSTGRES_PASSWORD = ""
POSTGRES_USER = ""
POSTGRES_HOST = ""
POSTGRES_DB_NAME = ""
POSTGRES_PORT = ""

S3_URL = ""
S3_USERNAME = ""
S3_PASSWORD = ""
S3_BUCKET = ""

LANGCHAIN_API_KEY = ""
LANGCHAIN_TRACING_V2 = ""
LANGCHAIN_PROJECT = ""

PGVECTOR_COLLECTION = ""
DATABASE_URL= "postgresql+psycopg://..."

Start the flask server

python main.py

Note

• The script will automatically generate some temporary files and keep on overwriting them. Currently index.html, output.csv and temp.wav will be the temp files.

• For current version, the ai generated posts that will be saved to database consist of the some hardcoded fields for now like the author_id, image_id, status, and type.

Test

curl -X POST http://localhost:8080/analyze-audio \\
    -F "audio_file=./audio_name.wav"