Get a summary of a youtube video using AI. Check the live demo.
Do you need a YouTube Data API v3 key & a Generative Language API key.
Create an environment file cp .env.example .env.local and set the following variables:
- NEXT_PUBLIC_GOOGLE_API_KEY is the API key for Google (Generative Language API).
- NEXT_PUBLIC_YOUTUBE_API_KEY is the API key for Google (YouTube Data API v3).
Run the development server: npm run dev
demo.mp4
- limit the AI request (Google Gemini API)
- select the AI model
- use only free AI models (??)
- prevent (by IP?) multiple requests to the AI
- switch between AI models & providers (Google, OpenAI)
- restrict the amount of chars to be processed by the AI ($)
- make more responsive
- change style & add instructions
- add related/more video links
- add getparam to work (video ID as query param)
- show more info about the video
- add 3 more options: 3 most important points, 5 most important points & a custom prompt (in a modal)
- sort voices by language
- play/stop status (when finish playing, it should change state)
- Voice in mobile (when no voice found)