Grow and Expand Your Podcast Audience with AI
The future of podcasting is short form. Don’t believe me? Ask the 144 million Americans that listen to one every year. Look at any of the experts telling you to diversify your output and create a specific feed for your “best of” segments. It all confirms what you already suspect: attention spans are getting shorter.
This isn’t to say that long-form podcasting is dying, but the medium is shifting. How is a producer or host expected to keep up with this demand? The big podcasting companies can hire interns to clip great moments from episodes (I used to be one of them), but what if you have a back catalog? What if your show is 3 hours long? What if you’re not even sure if your audience would like this kind of thing? It would be pretty costly time and money wise to even attempt unless you had some kind of superhuman power.
Fortunately, we live in 2020. Artificial intelligence and machine learning are poised to reshape nearly every digital landscape imaginable. I recently attended Podcast Movement Evolutions in Los Angeles and learned of some up and coming services that plan to utilize that exact technology to at least solve the problems I mentioned earlier.
One such service is AudioBurst. They’re trying to connect you with the 32% of the US population that listens to podcasts even if they prefer short content and you provide long content.
Their service uses a 4-step process to promote your audio.
First, they ingest. This means that upon signing up for their service you would then upload your audio to them in lieu of or in addition to your hosting service.
Then they index your audio. They deliver your audio as uploaded to your regular platforms, but they also scan and transcribe all of the audio from your podcasts. This in and of itself would be a great feature, but they don’t stop there.
After your audio has been indexed and transcribed, their machine learning software scans your audio for topics and patterns. It then segments these conversations based on the topics being discussed. For instance, if you were to talk about politics for 5 minutes of your 3-hour history podcast, those 5 minutes would appear as a singular segment.
Then finally, they deliver this audio to you for sharing. This eliminates potentially hours of sifting through and chopping up audio and gives you the opportunity to share those bits of specialized content to audiences who might be interested in just that.
When a podcast listener downloads AudioBurst, they get a curated feed just like you’d expect from Twitter or Facebook. Unlike those platforms though, the content being delivered would be audio personalized to their interests. This technology makes it infinitely easier to promote your podcast to people who would be interested, not by shows in their entirety, but my specific topics.
Another company is using similar technology to accomplish another difficult feat. When thinking of audio versus video, each has benefits and each has downfalls. Audio is great for multitasking since listening is a passive experience, but since video takes more of your attention consuming it, it is much easily branded as an interactive experience. A service called Adori is trying to bridge that gap.
Using a process very similar to the one listed above, they transcribe and index your uploaded audio. However, instead of clipping and sharing, they recommend interactive elements to their users.
That 5-minute politics discussion in your long-form history podcast? Perhaps it would be beneficial to poll your audience as to who’s opinion they like the most. In the past, this had to be done on a separate platform like Twitter, but any web strategist will tell you that every time you direct your audience to another platform, you lose about half of them in the process.
Adori would suggest a poll for that segment and upon listening your audience would see an opinion poll appear in the space where your podcast artwork used to be. This approach is also using AI in an innovative way to find the grey area between the passiveness of audio and the interactivity of video.
You can check out what that experience is like by going to the App Store for iOS and downloading their app.
As time marches on AI and machine learning will take over much of our lives. More and more every day the world’s best mechanics and surgeons are computers. Podcast producers have gotten complacent thinking that a computer could never do their jobs at the same level. However as the technology gets better and better, there is no reason to believe that soon the world’s best podcast producer and clipper could also be a laptop. As with previous technological advances, those who survive the revolutions tend not to be the resistance to change but the early adopters and with these new services’ low-entry points, you have few reasons not to join the ranks of the new AI podcasters.
Casey Franco has been producing podcasts for years, with previous experience with iHeartRadio and Stitcher, before becoming the Head of Production for Jam Street Media. He’s a futurist and former podcaster as well. Contact him by email at Casey@JamStreetMedia.com.