Show HN: We made a podcast search engine with 12M podcast transcripts

Y	Hacker News new \| ask \| show \| jobs

	Show HN: We made a podcast search engine with 12M podcast transcripts (fluiddata.com)
	32 points by nibbleshift 2774 days ago

3 comments

I really like your interface. It's easy to use and I was able to find some podcasts relevant to my queries. Bookmarked.

Thanks for the feedback.

How have you done transcripts?

We have a cluster that that is automatically transcribing newly published episodes.

What software do you use to actually generate the transcript from the audio?

We developed and built a distributed processing system that we use to automatically transcribe the audio.

Any plans for non-english podcasts?

We do have plans to support additional languages and we should have support for additional languages next year.