This is amazing. I fed it a Linux Bash shell & CLI reference guide in PDF format I had on my machine. It took about 10 minutes. But wow. Obviously it didn't go into any details. But it kinda gave a great overview of what bash is , how it works and how bash scripts can be useful.
NotebookLM's automatically generated podcasts are surprisingly effective
431–440 of 511 posts
Re: NotebookLM's automatically generated podcasts are surprisingly effective
#432Earlier quoted context omitted.
I’m not sure. The bestsellers lists are full of average-or-slightly-above-average wordsmiths with a good idea, the time and stamina to write a novel and risk it failing, someone who was willing to take a chance on them, and a bit of luck. The majority of human creative output is not exceptional. A decent LLM can just keep going. Time and stamina are effectively unlimited, and an LLM can just keep rolling its 100 dice…
I get your point, but using the best-sellers list as a proving point isn't exactly a slam-dunk. What's that saying? "Nobody ever went broke overestimating the poor taste of the average person"
I’m just saying, the vast majority of human creative endeavours are not exceptional. The bar for AI is not Tolkien or Dickens, it’s Grisham and Clancy.
Re: NotebookLM's automatically generated podcasts are surprisingly effective
#433Earlier quoted context omitted.
Must have been prompted to be an American podcaster. Bring on the one that's all British and snarky!
Might be hard to spot if you’re American in the US, but LLMs feel very American even outside of region and language. Concretely, I have to constantly ask for recipes to be changed to metric. Less concretely, the undertones and mannerisms of politeness, positivity and “excitedness” comes across as very American to me, probably even within the rest of the Anglosphere. How would I describe it? Maybe similar to how you’d…
As an American, I find it exhausting. I think of it as fake Silicon Valley/SF "nice" affect combined with non-US English as a second language floridity. My setup prompt for ChatGPT includes a reminder that if it answers too long or slowly, then it will take time away from my medical research grad students and people will probably die as a result of the delays. It helps a little.
Re: NotebookLM's automatically generated podcasts are surprisingly effective
#434One of my favorite ChatGPT uses is voice chat during long drives as a pseudo, albeit interactive, podcast to learn about various technical topics at the edge of my knowledge base. This podcast generation is pretty amazing, but hopefully they make the "competency level" of the hosts tunable. One thing I love is being able to guide ChatGPT to the technical level I'm looking for. Maybe I'm just bad at finding podcasts,…
Lex Fridman gets the best guests and asks the worst questions. I’d love to be able to tune up the technical level of those interviews…
Re: NotebookLM's automatically generated podcasts are surprisingly effective
#435I don’t think this is all that impressive, the generated podcast is pretty shallow - lots of ‘whoa meta’ and the word ‘like’ thrown into every sentence. Yes, it will generate a middle-of-the-road waffling podcast, but not one with any real depth.
To me, that's just how they tuned the 'audience' of the podcast, which I think we can imagine was at least partly informed by the 'audience' IRL podcasts are named at. I, too would like to be able to 'turn up the technical' on these, but for example, I dumped a paper about a latchless mutexless work distribution algorithm into it, which I had read but still had questions about, and the podcast accurately summarized, simplified it, and got my questions answered, which I then validated later by re-reading the paper. It was faster than combing through the paper would have been.
Re: NotebookLM's automatically generated podcasts are surprisingly effective
#436Re: NotebookLM's automatically generated podcasts are surprisingly effective
#437Re: NotebookLM's automatically generated podcasts are surprisingly effective
#438Earlier quoted context omitted.
go is a game with an obvious score function which can be used to construct a loss, well defined moves, and total visibility of the board. It is less obvious how to write a score for creativity in art or music, nor does it have well defined bounds on what is considered a legitimate construct of either. Just because computing hardware lets you multiple matrices faster does not mean we have the means to solve all proble…
Completely backwards. There is no obvious score function for Go. That's how AlphaGo broke through, it was able to figure out a scoring method to actually accurately gauge how well it was doing so it could learn and improve.
If what you are saying is true, how does anyone know the game is over? there is a clear win condition. I know that just like with chess, knowing the current vAlue of the board is difficult, but win or loss is clear.
Re: NotebookLM's automatically generated podcasts are surprisingly effective
#439The Deep Dive Podcast generator is amazing. Astounding even. I found out about it today and generated a couple. I generated on using a 38 page long PDF however and the 40 minute podcast it generated was awesome, but at 20 minutes (halfway thru) the conversation was mostly a repeat of things that were already said even though there was much more very important content that they omitted.
So it works great but just needs a bit of work to be done to cleanup things like that repetition. I wondered if this happened because there was a big "Table of Contents" in the doc, and maybe that made it see everything twice? I didn't try it again with a document lacking the ToC.
Re: NotebookLM's automatically generated podcasts are surprisingly effective
#440I actually find this alternative Google pdf-to-podcast service much better — it is less sensationalist and goes into more technical depth: https://illuminate.google.com/home?pli=1 Currently only handles arxiv PDFs.
I didn't know about this, thanks! I just tried one; to my ears it is much less engaging vocally. I like the extra detail though. I'd probably personally prefer to tune NotebookLM's technicality of audience if I could. I'd also like different speakers - I wouldn't be surprised if we all will recognize the NotebookLM speakers instantly in a year or so.