

However, we realized that using these screen-based tools can kill so much fun since these tools often divert the players' attention to the electronic device, which can distract them from the party with incoming messages and other notifications. There are plenty of web-based or app-based tools for truths and dares, just like for many other party games. We propose this small hand-size gadget with a built-in thermal printer for generating each prompt. For a simple purpose as generating truth or dare prompts, Markovify is a better choice. Markovify, as a lighter model, is environmentally friendly and easier to use. And due to the unpredictable nature of machine learning, there will still be a lot of uncertainty for the result even with a smarter model.

The heavier model also takes longer to train and run, which would be a nightmare for a game tool. Moreover, the pre-trained ML models are rather complex and heavy, meaning they're computationally expensive and requires a better machine. The absurdity of such primitive ML application and the absurdity of the Truth or Dare game itself are just perfect match - "errors" in machine learning can be fun when put into the right place. Compared to pre-trained text generation models like GPT-3, Markov Chain models actually have a very poor understanding of the input text, which means the generated results make less sense. In the text generating process, a Markov chain model predicts the next word based on the previous word. Work in progress: assembling Why Markov Chain?Ī Markov chain is a stochastic model that generates random results based on probability.
Truth or dare generator series#
By using a series of selected inputs, including both absurd and serious content, Morkovify can generate 'infinite' challenges for the players that are beyond the human imagination. The fact that this machine learning model doesn't really understand the true meaning of each sentence results in the absurdity of the prompts, and makes it more unpredictable. Markovify predicts the next word based upon its previous word, but not the whole sentence.
Truth or dare generator generator#
We propose an innovative way to generate prompts using Markovify - a simple machine learning generator based on Markov Chain.

It is challenging to continuously devise clever prompts yourself and the prompts you can find online are often limited and repetitive in content. However, r unning out of prompts is a common pain point of the game. Truth or Dare is a classic party game, which places the player in a dilemma between answering a tricky question or performing a challenging task. Trouble Maker is a game-gadget that generates random prompts for Truth or Dare using Markov Chain. Results Announced for Community Choice Prize See All Winners Trouble Maker: A machine learning based Truth or Dare generator
