search

lord-denning / Huberman-Lab-Podcast-Transcripts

247 stars 54 forks source link

Questions about methodology to generate these transcripts #2

Open jvican opened 2 years ago

jvican commented 2 years ago

@lord-denning Do you use some tooling to query the Youtube API and access the Youtube transcript of the videos? Or is the process completely manual? Curious to hear about what your process is like.

lord-denning commented 1 year ago

Hi, sorry for the late reply. Don't sign into this account often. I use the youtube transcript and a few word macros to help with formatting.

Otherwise the process is completely manual (I'm a lawyer and don't know how to code). That's why I don't have time to continue doing this. It takes between 30 mins to an hour to transcribe each episode.

jroge commented 1 year ago

@jvican I think you may be able to - I found this on stack overflow https://stackoverflow.com/questions/14061195/how-to-get-transcript-in-youtube-api-v3. I may try to see if I can build this out in a few months just to see if I can keep this repo going

lukerob5 commented 1 year ago

I made a gist for generating transcripts of YouTube videos that could be used for this, if anyone is interested (requires a little bit of python knowledge to run). https://gist.github.com/lukerob5/7d8e30edcaa0bb578e76e3f26917c91c

prakhar625 commented 1 year ago

Hi all, found a faster and better way to get all the transcripts. You can find all the transcripts here on my repo: https://github.com/prakhar625/huberman-podcasts-transcripts

Let me know if anyone can help me format the text files in a better way - programmatically.

jvican commented 1 year ago

Huberman Transcripts has now a website, it uses OpenAI's whisper models and it also does summarization and support search (by terms). It's been my go-to for the past week. https://www.hubermantranscripts.com Highly recommended