Generative AI
Particularly if all it takes is 1 click on to show any content material into podcast
For many who might not know, NotebookLM is a personalised AI analysis assistant powered by Gemini 1.5 Professional, designed to make sense of complicated data. Along with answering questions based mostly in your uploaded sources (paperwork, slides, charts, and so forth.), it could additionally create personalised examine supplies by robotically producing issues like a desk of contents, examine guides, briefing paperwork, FAQs, and extra. Whereas it formulates solutions based mostly on the uploaded sources, it additionally supplies inline citations, highlighting the precise textual content blocks within the supply paperwork used to generate the response.
The uploaded content material can vary from analysis papers and assembly transcripts to quotes from fascinating books, chapters of a novel you’re writing, company paperwork, and extra. These sources can embody Google Docs, Slides, PDFs, textual content recordsdata, copied textual content, and even internet pages.
Now, to the primary purpose for this text: Final month, NotebookLM introduced a brand new characteristic — Audio Overviews — which has been making headlines. This characteristic affords a brand new method to work together along with your supply paperwork. With only one click on, it generates partaking “deep dive” discussions that summarize the important thing subjects in your sources.
What’s much more spectacular is the way it transforms any piece of content material, irrespective of how dry, by producing two AI hosts (one male and one feminine) who talk about the doc’s contents in a podcast-style format.
For those who’re questioning what “podcast-style format” means, think about the pleasant banter, the little jokes, the back-and-forth conversations, the laughs, interruptions, “umms,” and “you is aware of’”— basically all of the hallmarks of an ideal podcast listening expertise.
These podcast-style conversations create pure connections and segues out of your textual content, leading to an enticing dialogue.
To check it out, I made a decision to repurpose certainly one of my outdated Medium articles and create a podcast from it to cater to a extra audio-loving viewers.
The arrange for a similar was fairly easy.
- Go to NotebookLM. You’ll need to check in along with your Google ID in the event you aren’t already. If it’s your first go to, you’ll see a number of pattern notebooks and you’ll create a brand new one with the “Create” button.
- Subsequent, add content material to your pocket book. I used the web site supply to feed in my Medium article. Alternatively, you’ll be able to paste textual content or fetch from Google Drive.
- Lastly, click on the “Generate” button contained in the Pocket book information (see picture beneath) to create the audio. And go seize a ☕️ as it would take a couple of minutes relying on the content material size.
P.S. It took round 4 minutes to generate a 13 minute audio from my 1100-word article. You’ll be able to play and pay attention right here.
P.S. I ended up attempting Audio Overview with numerous sources, resembling podcast transcripts, analysis papers, and knowledge science blogs. The next takeaways are an amalgamation of my experiences throughout all these sources.
Let’s begin with the great things:
- It’s outstanding that we will shortly create a podcast episode in simply minutes, permitting many people to have a facet gig as podcasters (must you select to). It is a wonderful means for writers to repurpose their content material and for others to have interaction with comparatively complicated subjects in a enjoyable and accessible method.
- The usage of analogies all through the audio is actually outstanding and fascinating. Within the case of my Medium article, it was in a position to take a comparatively area of interest (learn:boring) matter (scaling challenges with Gen AI won’t attraction to everybody exterior the instant discipline) and make connections to on a regular basis issues.
As an illustration, at one level the hosts talk about Gen AI token prices and supply a way more relatable instance, evaluating how these prices can add as much as micro-transactions in a cell sport. Equally, they clarify immediate engineering with an instance of offering an entire recipe with measurements, moderately than merely saying “make me a scrumptious meal”. In addition they use the analogy of a automobile remembering a standard route to clarify LLM caching. - The best way the 2 hosts construct on one another’s sentences feels very pure, and the segues circulation seamlessly. For instance, utilizing phrases like “talking of…” to introduce a brand new matter feels natural and never pressured in any respect.
- Emphasis on sure phrases at simply the suitable moments helps maintain the listeners’ consideration. Expressions like “oh wow”, “oops”, and “aah” convey real shock at what the opposite host simply mentioned. Pure pauses to think about the suitable phrase make the dialog really feel spontaneous moderately than rehearsed.
- After testing this on a number of deep studying papers, I can confidently say it is going to be a sport changer for explaining complicated analysis that advantages from analogies and “clarify like I’m 5” (ELI5) examples. In actual fact, the rules in certainly one of their pre-prepared instance notebooks, titled Introduction to NotebookLM, state that it’s designed for researchers, journalists, college students, and enterprise professionals.
Having appeared on the key benefits, there are additionally just a few limitations to think about:
- Typically, the dialog between the 2 hosts doesn’t really feel actual. Fairly often, they end one another’s sentences, even when the primary host has simply requested the second host to clarify a brand new idea and some seconds later, Host 1 finally ends up answering their very own query.
- Not all enter sources generate audio of equal high quality. As a part of stress testing, I attempted inputting the transcript from one other podcast, and the hosts appeared extra inclined to make humorous noises at one another — ‘yayaya,’ ‘oh yeah,’ ‘hmm,’ ‘uh-huh,’ ‘proper,’ ‘gotcha,’ and so forth.!
- The one draw back to having a variety of analogies whereas discussing a subject is that generally the AI can get the analogies mistaken. As an illustration, whereas discussing a weblog on forecasting metrics, it used the analogy of “similar to in faculties a decrease rating is mostly higher, it means your forecast is nearer to actuality”.
Such hallucinations are frequent throughout completely different generative AI fashions and have been included as a disclaimer of their instrument as properly. These is likely to be extra pronounced if we offer a really area of interest, extremely specialised matter, such because the position of microRNAs in gene regulation (the subject that received the Nobel Prize in 2024 this week). In such circumstances, it might begin hallucinating with analogies used because of an absence of related inherent information🤷♀. - For very massive texts, the podcast can usually finish abruptly. This means that there could also be a cutoff level for the coaching knowledge, past which the audio can not adapt to supply a clean, pure ending.
- (Very minor however) A number of the phrases, principally abbreviations, are garbled within the audio. For some purpose RAG is pronounced as ArrrR-G as an alternative of particular person alphabets like R-A-G.
- At instances, hosts overly agree with each other, utilizing filler phrases like ‘proper’ and ‘precisely’ whereas the opposite host remains to be speaking. This will really feel like pressured responses; I imply, let the poor man end!
Now that we’ve coated the nice and the dangerous, let’s transfer on to the million-dollar query: is that this new tech sufficient to offer podcasters a critical competitors?
My easy reply is — not but. The rationale? All of the aforementioned points we’ve mentioned. And I do know a few of you would possibly disagree and say these issues are minor, and also you’d be proper. For those who take heed to only one podcast, chances are you’ll not even discover them, however in the event you repeatedly take heed to a number of episodes, particularly on a each day or weekly foundation, the sheer variety of analogies and “exactlys” can change into overwhelming. For these causes, maybe Google by no means positioned it as a podcasting instrument of their preliminary launch.
That mentioned, it can positively decrease the barrier to entry for a lot of who need to discover this discipline however might not need to use their very own voice for numerous causes. Extra importantly, I see its use as a method to eat complicated subjects in digestible codecs.