Five product ideas on Music, Content, and News — improving discovery & personalization
A list of product ideas — from my personal backlog — which could trigger interesting entrepreneurial discussions.
Music & content discovery, smart book annotations, and personalized video news synopsis
The following list briefly describes selected, product concepts — from my personal backlog — which could trigger interesting entrepreneurial discussions — feel free to comment, use/ extend, or build on top.
1. Digital comments on Physical books!
You are reading a physical book — novel, technical or other. You are at an important point/ paragraph where you need help or you want to add a comment, take a note, or ask a question. What if you could use your smartphone to seamlessly submit your notes or discover popular notes from other users — on the particular paragraph you are currently reading?
This is how it could work:
You point your smartphone on a paragraph or a phrase in the physical book. Then the ‘book annotation app’ instantly performs the following steps:
- Extracts the paragraph/ part of the page you are pointing the device— by using OCR (optical character recognition)
- Uses the extracted paragraph/ text to perform a full-text-search against a large database of books — via a service call to Google’s Book API or similar.
- Receives the response from the API — including the identifier of the book, metadata about the book, and positioning information — a reference to the paragraph and page where the match was found.
- Retrieves user-generated-content and metadata about the specific paragraph/phrase/page of the identified book.
- Presents the summarized user-generated content to the user via the app — possibly in an Augmented Reality mode and/ or with voice support.
- The user interacts with voice — to explore user-generated content and/ or append private or public comments on the identified paragraph of the book.
Alternatively, step 1 above, could be omitted, if the app fully supports Voice-driven interaction: instead of scanning the page and doing OCR to retrieve the text, the user could ask his/her Digital Assistant: ‘Google/Alexa/Siri/Cortana, tell me more about the paragraph starting with {Read the start of the paragraph}’. Then, by applying speech to text, the process moves on to step 2.
2. Discover music, loved by people around you!
A new music discovery option — based on the popular & recently-played songs from people in proximity to you!
Imagine if you could discover music that is popular in the groups of people you join or interact with. For example, you could discover music which is popular among the members of the concert audience you just joined (where you could expect some similarity in terms of music preferences) or in a theater, a university classroom, or in the train (where the music preferences could have very interesting diversity and increased ‘discovery value’).
The ‘social music discovery’ app could share ‘anonymous music statistics and preferences’ (for example, via music streaming apps) and the ‘current location’ of the user. The app could then derive the most frequent patterns and music preferences from people in proximity (at the moment or over a time frame). Users in this location-defined group could then browse/ discover music, which is popular within the group (or a previous instance of it).
This could empower an ongoing, social-driven music discovery engine (influenced both by people I know and/or strangers which happened to be in the same place/ location).
3. News, trending around you!
A mobile app, able to instantly capture the signals regarding content/news consumed by people around you.
This app could make intelligent recommendations based on popular content, dynamics, and trends in reference to a specific social arrangement. Imagine if you could browse the most read articles among the people in your company building, or those strangers in the train, or the audience of a movie you are watching; or an academic class. This would be extremely helpful not only as a news/content discovery engine but also as a helper for the identification of common interests and conversation topics. Could be an interesting discovery function for the Medium app
4. Public places adapting their music to the current audience!
Imagine a restaurant or cafe which can seamlessly ‘understand’ the music preferences of its current audience and then serve them the right music — a sequence of songs with higher probability to satisfy the majority of the audience: a just-in-time synthesis of music, matching the stated or implied preferences of the current audience — set of customers in the physical space.
This has an interesting extension allowing the discovery of places (for instance restaurants, cafe, bars, etc.) based on the aggregated musical preferences of the people frequently visiting the place (customers, visitors, etc.); users could search and explore places with all traditional criteria and also the music preferences of the ‘typical customer’.
5. ‘Personalized News Synopsis’ via interactive videos
Personalization on top of collections of independent pieces of content is quite conventional nowadays. But how about dynamically compiling a short news video that is personalized for each user? Imagine an intelligent process to summarize the daily news as one video, which is also interactive and according to the user’s implied and/or stated preferences.
The generation of the content synopsis can be done using both human editors and advanced Artificial Intelligence and content understanding technologies. The daily synopsis process, outputs a large set of small pieces of ‘video stories’, which are properly tagged (for discovery and personalization) and packaged (to allow smooth matching with other video-stories, when dynamically combined within the ‘personalized synopsis video’ for the particular user)
Each ‘video story’ could be presented by a human newscaster or an avatar - using AI — synthesized voice and auto-video generation technologies. The structure of the ‘News Synopsis’ would allow smooth and interactive navigation through the video, either explicitly (the user is clicking ‘next story’) or implicitly (the system is filtering out or re-prioritizing stories within the video, based on the user’s profile, preferences, session details, day of the week, current user engagement metrics, etc.).
