A Search Engine for Spoken Words in YouTube Videos

Podzinger is a Speech-to-text search engine that allows us to search the audio contents of YouTube videos for spoken words. The engine is powered by 30 year research on speech recognition by BBN Technologies, and allegedly the speech recognition technology is used by the CIA amongst other government agencies. In any case it’s pretty impressive.

A search for a “series of tubes” shows the text it recognized from the video, along with a marker for where the sentence in uttered in the video. Oddly enough, it translates “tubes” as “two” but still manages to get find the clip I was looking for.

Try it out, you can also search for podcasts.

By the way, if for some reason you haven’t heard of the phrase “the internet is a series of tubes” — watch this!.

No Comments, Comment or Ping

Reply to “A Search Engine for Spoken Words in YouTube Videos”

Please read the Terms of Use before commenting!

Basic HTML allowed (a, blockquote, strong, em)