Mythical AI

Mythical AI Substack explores the practical applications and future possibilities of artificial intelligence. It addresses training AI models on personal data, agent AIs, AI in programming and content creation, digital cloning, reverse engineering prompts from images, innovative AI tools, speech to text technology, and character consistency in image generation.

Artificial Intelligence Applications AI Model Training Agent AIs AI and Programming Digital Cloning Reverse Engineering Prompts Innovative AI Tools Speech to Text Technology Image Generation

The hottest Substack posts of Mythical AI

And their main takeaways
235 implied HN points 19 Feb 23
  1. Large language models like ChatGPT can summarize articles, write stories, and engage in conversations.
  2. To train ChatGPT on your own text, you can use methods like giving the AI data in the prompt, fine-tuning a GPT3 model, using a paid service, or using an embedding database.
  3. Interesting use cases for training GPT3 on your own data include personalized email generators, chatting in the style of famous authors, creating blog posts, chatting with an author or book, and customer service applications.
98 implied HN points 12 Jun 23
  1. Creating a digital clone with AI involves capturing content, voice, likeness, and movement.
  2. Potential problems with AI clones include consent and ownership issues, problematic responses, and scams.
  3. There are tools available to create parts of an AI clone, but building a full AI clone is a complex process requiring integration of various technologies.
137 implied HN points 07 Apr 23
  1. AI is making it easier for people to program by allowing them to describe tasks in English and having the computer figure out the code.
  2. Computers need precise instructions and struggle with understanding context, making programming challenging.
  3. Programmers are rare, expensive, and building software is costly, but AI is helping automate coding, making programmers more productive.
98 implied HN points 24 Mar 23
  1. Creating videos from text prompts is challenging because it involves understanding and replicating movement besides images.
  2. Existing text to image systems are amazing but doing text to video requires additional capabilities.
  3. While there are research papers and tools for text to video, there's no high-quality solution yet, but advancements are expected in the future.
Get a weekly roundup of the best Substack posts, by hacker news affinity:
58 implied HN points 17 Mar 23
  1. AI tools are making our lives easier and more fun, from simple toys to advanced APIs.
  2. Various AI tools are available for different tasks like running prompts on different models, converting paper forms to computer text, and summarizing videos with automatic chapter markers.
  3. Advancements in AI technology are leading to the development of innovative tools like text-to-song generators, text-to-vector image converters, and color book page generators.
19 implied HN points 08 Mar 23
  1. Speech to text technology has a long history of development, evolving from early systems in the 1950s to today's advanced AI models.
  2. The process of converting speech to text involves recording audio, breaking it down into sound chunks, and using algorithms to predict words from those chunks.
  3. Speech to text models are evaluated based on metrics like Word Error Rate (WER), Perplexity, and Word Confusion Networks (WCNs) to measure accuracy and performance.
19 implied HN points 10 Mar 23
  1. The post covers the best speech to text apps you can try today like Apple Dictation, Otter.ai, and Descript.
  2. It provides an overview of free open-source speech to text models you can use, like Whisper and Vosk.
  3. The post also lists paid speech to text APIs, such as Deepgram, AssemblyAI, and Google Speech-to-Text, with their pricing and features.