Podcast content is growing each year, with over 4.5 million active podcasts globally — and nearly 30 million episodes published in 2023 alone — the challenge of discovering relevant content has become increasingly complex. Podyssey emerged as a solution to this challenge, with its AI-driven approach to podcast curation and personalization. Ahmad Saleem, co-founder and CEO of Podyssey aims to set new standards for how listeners find and engage with relevant podcast content.

"In the 1980s and 90s, the antidote to a mid-life crisis was starting a suburban band. Now it is starting a podcast.”

Historically, the only way to find relevant podcast content was by manually searching through titles of shows and episode descriptions, relying heavily on ratings, reviews, and basic categorization. But these traditional discovery methods have proven insufficient as the volume of content continues to increase. Listeners often find themselves overwhelmed by choices or stuck in content bubbles, missing out on potentially valuable podcasts that fall outside of their usual listening patterns.

“As a listener, the task of shifting through this new content is onerous and daunting. In fact, just finding that content you already like is becoming more and more daunting,” comments Saleem. “This is largely because there is more content to sift through; but also because the method for doing so is increasingly outdated.”

Podyssey's approach puts curated content discovery at the forefront. At its heart lies a powerful and unique search engine that leverages cutting-edge technology to understand the actual discussions within podcasts. This isn't just an incremental improvement – it's a complete paradigm shift in how listeners find and consume podcast content.

"Traditional platforms have to rely on the genre or content classifications applied by the podcaster to their content," explains Saleem.

“This approach might have worked in simpler times, but today's podcast landscape defies such basic categorization. Take Joe Rogan's podcast, for instance – trying to pin it down to a single genre is like trying to fit a square peg in a round hole. The reality is that podcast content has become more genre-fluid than ever before, particularly in long-form formats.”

The technical innovation behind Podyssey is substantial. The platform employs a sophisticated combination of audio recognition algorithms and natural language processing, utilizing bi-encoders and cross-encoders to analyze user queries. This goes beyond simple keyword matching, allowing the system to understand the intent behind searches and deliver more relevant results, even without exact matches.

"There is nothing else like Podyssey out there," according to Saleem. "We are the only podcast platform that allows listeners to curate their listening based on the content being discussed." This means listeners can finally break free from the constraints of genre-based discovery.

Podyssey goes beyond search capabilities. By studying listener behavior – from search patterns to engagement rates with specific clips – the platform creates highly detailed user profiles that enable complete personalization. Instead of sitting through a three-hour episode for ten minutes of relevant content, users can enjoy sixty minutes of carefully curated material drawn from various sources, all focused on their specific interests.

The company’s journey is about more than just improving podcast discovery – it's about reimagining the entire podcast ecosystem. While traditional platforms have remained relatively unchanged since the industry's inception, Podyssey represents a fundamental shift in how both creators and listeners engage with podcast content. By removing artificial boundaries and putting content relationships at the center of the experience, Podyssey isn't just participating in the podcast revolution – it's leading it.

The message is clear: the future of podcast discovery has arrived, and it's time for listeners to break free from the constraints of traditional platforms. Podyssey stands ready to ensure that every listener can find exactly what they're looking for – and discover content they never knew they needed.

What challenges do listeners face in discovering new podcasts in today's oversaturated market?

Like most other types of media, podcasts are suffering a crisis of content oversaturation - a problem created by the sheer amount of new podcast content being produced. There were roughly 30 million podcast episodes published in 2023. And this trend does not look like it will slow down anytime soon. In the 1980s and 90s, the antidote to a mid-life crisis was starting a suburban band. Now it is starting a podcast. It is a cultural trend that is here to stay.

As a listener, the task of shifting through this new content is onerous and daunting. In fact, just finding that content you already like is becoming more and more daunting. This is largely because there is more content to sift through; but also because the method for doing so is increasingly outdated. Up until Podyssey was released, the only way to find podcast content was searching the titles and show notes of episodes. This may have worked when there was not that much content around and when what was being discussed within the content was simpler (i.e. people created an episode around discussing one specific topic). Both of those conditions have changed. We are producing far more content now. And it is long form discussions weaving in and out of various topics. The original system of searching cannot cope anymore. That’s why we created Podyssey.

**How does Podyssey’s AI technology differ from traditional podcast recommendation systems? \ There is nothing else like Podyssey out there. We are the only podcast platform that allows listeners to curate their listening based on the content being discussed. Traditional platforms can only do it at the genre, podcast or episode level.

Traditional podcast platforms start curating content at the genre or podcast level. They have to rely on the genre or content classifications applied by the podcaster to their content. Traditional platforms use this to try to figure out what genres you like or what podcasts you are already listening to, and based on those inputs, recommend other content to you. An example would be that if you liked a sports podcast, then they might recommend another sports genre podcast.

The above approach has worked well but there have always been issues with it. The foremost one is that some podcasts are hard to pin down into one genre - for example, take Joe Rogan, it is not easy to put his podcast in any single genre. Its content is too diverse. Overall, a lot of podcast content is becoming more genre fluid than before - particularly long form content. Segmenting content at such a basic level as genre is not helping that much in content discovery anymore.

Podyssey is a new type of podcast platform, which at its heart is powered by a powerful and unique search engine. The engine uses the latest in language processing to better understand the content being discussed on podcasts and uses that to assign content into various subject and topic buckets. This way, the search capability allows listeners to find content that is most interesting to them - irrespective of what genre, podcast or episode it falls into. On Podyssey, we are no longer bound by any of the existing constraints. Want to search for where on an episode Joe Rogan talks about US Politics, we will find that content for you. We are putting the content and its relationship to the listener at the heart of our platform.

**Can you explain the process of how Podyssey’s AI creates a personalized profile for each user/ how it cuts through the noise to identify relevant, personalized content? \ By studying the type of content listeners are searching for and listening to, we can create a much more specific profile for a listener. This allows us to be able to recommend content at a level that is just not possible on any other platforms. And by focusing on specific clips within episodes that relate to the content that a listener finds interesting, we cut through the noise of what a listener does not want to listen to. No need to listen to a 3 hr Joe Rogan episode. Just find the parts where he talks about the topic you are interested in.

**What does Podyssey consider when curating content for its users? \ Podyssey’s unique offering is that it allows users to curate the content they are most interested in listening to. If you are interested in a topic, then we will find content that matches that topic in any podcast or episode and play it for you. Extending that further, we can then curate content to that specific topic. Rather than listening to an hour long podcast episode which only talks about your topic of interest for 10 mins, you can listen to 60 mins of content on the topic you like curated from various different sources. In that curation, if there is content you do not like, then just skip that part and keep going. We will remember your preferences in what you like and what you skip and continually keep curating the content for you. Effectively Podyssey will get to a point where you can create your own podcast on a specific topic.

What specific AI techniques does Podyssey employ to understand user interests and preferences?

Podyssey uses audio recognition algorithms and natural language processing algorithms to handle the podcast content with it. These drive the search engine. The user interests and preferences are handled by a recommendation engine that utilises various inputs including but not limited to, user search history, their engagement rate on the various clips, etc.

Here’s a bit more technical answer: For our search engine, we use a combination of bi-encoders and cross encoders (both modules use sentence-transformers) to analyze a user's search query. This goes beyond keyword searches, and allows us to understand the intent behind the queries. Thanks to this, we are able to fetch more relevant search results, even if there are no exact keyword matches.

Our Landing Page and Discover section employs collaborative and content-based filtering and behavioral patterns to understand what users are listening to. As mentioned in above answers, we keep track of various engagement events like user searches, their likes, engagement on a clip, shares, follows etc. Apart from these, we ask user’s favorite categories and podcasts while onboarding them. All this information is fed to our recommendation engine, which finally outputs the set of clips to be recommended to the end users.

**How does Podyssey’s AI handle niche interests/ less mainstream podcast genres? \ I think this is where Podyssey shines - it allows you to dig really deep for the content you want to find. By looking at everything said in each podcast we can find any instance of a specific topic or interest. If it is out there, we will find it.

**How does Podyssey ensure user privacy while collecting data for personalization? \ We use Firebase Datastore to store sensitive personal information. Data privacy (and security) at rest is ensured by Firebase. The recommendation engine is the only system which pulls the personal data from Firestore. Firebase uses HTTPS to encrypt the data in transit, ensuring data privacy.

In addition to personal information, our recommendation engine relies on user activity for personalization. This activity data is stored on a separate server. Firebase generates a unique random ID for each user, which is then used to track their activity across the application. Importantly, our logging and monitoring servers never store any personal information—they only store the user IDs, ensuring that sensitive data remains protected while allowing effective personalization.

**Can you share any examples of when Podyssey has successfully matched its users with unexpected but enjoyable content? \ We had a user recently leave a review on the Apple Store which summarizes how Podyssey helps listeners find content.

The user was after a specific podcast snippet from a few years ago on mobile app development. They could not remember the name of the podcast, or which episode it was but they remembered what they were discussing. They searched for it on Podyssey. In a matter of seconds, they were able to find the exact content. But Podyssey was also able to make recommendations on content along the same lines that they were not even aware of. The result was that they discovered new podcasts that cover their topics of interest and are now part of their routine consumption.

**What are the potential future developments for Podyssey’s AI curation tech? \ Currently Podyssey allows you to find content by “what” is being discussed. Later this year, we hope to bring out a feature where a listener can discover content by “who” is talking. We are developing a process of automated recognition of speakers within podcasts. This will allow users to search by specific people discussing specific topics. This will be a game changer in the podcast world.

How do you anticipate Podyssey’s approach changing the podcast industry landscape in the coming years?

Podcasts are not a new phenomenon; they have been around for a few decades now. Most of the traditional podcasting platforms around today came into existence pretty much at the start of the podcasting industry. And if we are being critical, then those platforms have not changed considerably since they first came out. We envisage Podyssey being a new way of listeners consuming podcasts. And possibly a new way for podcasters to create content. Currently both, creators and listeners have to abide by rules that dictate how their content has to be categorised and how it can be stored in these traditional platforms. With Podyssey, we are re-imagining the whole podcast ecosystem.

There are no more boundaries holding content back.