Stable Diffusion to generate spectrograms to convert to sounds

December 16, 2022

Topic

Statistics / AI, music, spectrogram, Stable Diffusion

Stable Diffusion is an AI model that lets you enter text to generate images. Spectrograms visually represent sound. Seth Forsgren and Hayk Martiros combined the two for Riffusion, which lets you enter text and the model generates a spectrogram that is converted to audio.

Read more about the process here.

Projects by FlowingData See All →

Marital Status by Age

Separately, we looked at marrying age, divorce rates, and those who never married. Now let’s look at marital status all together, with the addition of the widowed status.

Data Underload #2

Merry Christmas, Bedford Falls! Merry Christmas you old Savings and Loan!

Cuisine Ingredients

What are the ingredients that make each cuisine? I looked at 40,000 recipes spanning 20 cuisines and 6,714 ingredients to see what makes food taste different.

Shifting Causes of Death

The most common causes of death changed over the years. They vary across sex and age group. This animation shows the details of these changes.