Generating Novel Cocktail Recipes with a Specific Style through Recurrent Neural Networks

Deep learning, a form of machine learning & AI, has been successfully utilized in a whole range of applications and is present in your smartphone as well your living room if you own a Google Home or Amazon Echo device. These smart speakers use natural language processing (NLP) to parse your speech into text and then analyze the text. Analogous technology can be used to generate text based on rules and associations learnt from a database of texts. And if your database consists of cocktail recipes and you’re interested in flavor as well as novelty, this could be used to generate new and uncommon cocktail recipes.

Using a recurrent neural network for NLP, this is exactly what I did. And as my database of ~500 cocktail recipes entirely stemmed from the famous NYC bar Death & Co, the resulting cocktail-AI learnt to generate recipes inspired by the style of Death & Co. I don’t want to tease you too much, so here’s the gist: cocktail recipes generated by the cocktail-AI are frequently suprising & surprisingly delicious! Depending on your interests (and level of stress) you might find either my technical account or the part where I try and share some of the generated cocktail recipes interesting. Cheers!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this:
search previous next tag category expand menu location phone mail time cart zoom edit close