AudioGPT πŸŽ™

PLUS: Free Prompt Engineering Course, AI Hits, and Additive Prompting

Hi folks!πŸ‘‹πŸ» This is The Prompt! We're your go-to source for all things AI!

On Monday, I’ll be shipping a curated list of 250+ AI directories where you can list your AI App. You can get up to 20k clicks to your app. If you want to get notified about the launch, sign up here.

The Prompt subscribers will get the list for free.πŸŽ‰

Now, back to the latest πŸ‘‡πŸ»

FEATURED

AudioGPT: Audio conversations πŸŽ™

AudioGPT is doing to audio what ChatGPT did to words.

It is a multi-modal system designed to understand and generate audio in spoken dialogue.

It can also do a bunch of other tasks like:

  • transcriptions from audio;

  • music and sounds from images;

  • talking head video from an audio file;

Try the model here, get the code here.

This model can only be used for non-commercial purposes for now.

WHAT ELSE IS GOING ON

πŸ’¬Β Apple enters the AI Chat. Sources say that Apple is working on a health coaching service named Quartz that will track and analyze your emotions.

πŸ’¬Β Meta wants to enter the AI Chat. In Meta’s earning call yesterday, Zuckerberg shared *vaguely* that they plan to integrate AI agents across all their apps.

πŸ‘€Β Harvey, the legal AI app raised $21M. Probably the fastest-growing AI app on the market right now. And they’re hiring aggressively.

πŸ‹πŸ»β€β™€οΈ Stability AI releases DeepFloyd. This model has high photorealism and language understanding. Yes, we’re getting closer to having good old English words on AI-generated photos. You can try to generate photos using Colab or access the code here.

RESOURCES

The best resources we came across lately that will help you become better at writing prompts & building AI apps.

πŸ“šΒ Harry Potter x Addidas AI-generated Collection [ Fun Twitter thread ]

πŸŽ₯Β Additive prompting framework with Midjourney [ a must-read thread ]

TOOLBOX

The latest AI tools to use or get inspiration from.

PROMPT OF THE DAY

TOOL

Midjourney

PROMPT

Beautiful {platform} icon for {describe app}. Concept {concrete objects}. Flat render, subtle gradients, no letters

RESULT

LATEST PAPERS

  • ChatVideo: A Tracklet-centric Multimodal and Versatile Video Understanding System

  • LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions

  • q2d: Turning Questions into Dialogs to Teach Models How to Search

  • mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality

  • Multi-Party Chat: Conversational Agents in Group Settings with Humans and Models