Demo

The App

https://tube-to-text-coach.streamlit.app/

Prompts

Prompts used in the app are publicly available on the LangChain Hub:

1. Short routine prompt
2. Long routine prompt

Local development

Requirements

Create a python virtual environment (I used python 3.10)
and install dependencies from requirements.txt

pip install -r requirements.txt

for yt dl to work locally you need to have ffmpeg installed

on mac

brew install ffmpeg

on ubuntu/debian

sudo apt-get install ffmpeg

on Windows

choco install ffmpeg

Log to Langsmith

Rename secrets.toml.dev to secrets.toml and set your Langchain API key

Launch the app

 streamlit run tube-to-text-coach.py