We use lots of AI to analyze your favorite content.

Here’s all the AI

It starts with an audio podcast or a YouTube video. We then point the AI at them to help you find and get fun insights from your content. See below for all the things we do.

Semantic Search uses AI to let you search by meaning. So if you search for “Holidays”, it will know Thanksgiving and Christmas are holidays.

Traditional Human Keyword search will do fuzzy search for the words you input.

Advanced Search Options

Human Keyword search has these features:

Phrase query: a specific sequence of terms that must be matched exactly. A phrase query needs to be surrounded by double quotes ("). For example, the query "The Pete and Sebastian Show" only returns a record if it contains the exact phrase “The Pete and Sebastian Show”. Fuzzy search is disabled with phrase query so you can narrow in on exactly what you’re looking for.

Prohibitor exclusion: excludes records that contain a specific term. To exclude a term, you need to prefix it with a minus (-). The engine only interprets the minus (-) as a prohibit operator when you place it at the start of a word. A minus (-) within double quotes (") is not treated as a prohibit operator.

Some examples:

  • pete -sebastian only matches records containing “pete”, but not “sebastian”

  • pete-sebastian matches records containing “pete” and “sebastian” (there’s no exclusion because the minus (-) is in the middle of the word)

  • -pete matches every record except those containing “pete”

  • -pete sebastian matches records containing “sebastian”, but not “pete”

  • "-pete" matches records containing “-pete” (no exclusion performed)

The AI identifies key People, Places and Things. You can click any to see all mentions across all episodes or just a specific episode

When shows have a lot of episodes, you can get a lot of AI keywords. From the show pages, you can see the top 25 most popular AI keywords

You can also bring up a page that shows ALL keywords for your content.

You’ll notice many times we take what was written or said by the humans and use AI to summarize

We use AI to show the emotion that’s part of each section of the episode

For YouTube videos, we load in the comments and use AI to determine the sentiment of each comment (positive, neutral or negative)

We also show the top positive and top negative comments.

The AI gets things wrong, so if you see an issue with the transcript you can notify us to fix it. In this example the AI transcribed “Jay Moore” instead of “Jay Mohr”.

For fun, we pass the episode text through the Flesch–Kincaid algorithm to see the reading level.

If you find a sift that you like, you can click play to hear just that snippet of audio and then link back to the source podcast or YouTube video.