JAKARTA - Spotify seems to be preparing a new move to make its algorithm more personal and, this time, more "obedient". The Swedish music streaming platform is reportedly developing a new feature called Notes that allows users to refine their Taste Profile through written input.
This information was first revealed through the decompilation of the Spotify app code version 9.1.28.385 by Android Authority. In the findings, there are a number of code strings that lead to the "Notes" feature that is directly connected to the Taste Profile - the main foundation of Spotify's recommendation system.
Until now, Spotify has built a Taste Profile based on user listening habits. Streaming data is the fuel for recommendations such as Discover Weekly, the home screen (Home), to annual summaries such as Spotify Wrapped and Blend.
In theory, this algorithm is precise. But in practice, recommendations sometimes go astray - for example, when users play a sleep playlist or white noise that actually "confuses" their main music preferences.
Currently, the only way to correct the direction of the algorithm is to exclude certain songs or playlists from the Taste Profile. Spotify says the change will take effect in 48 hours, so the content's influence on recommendations and taste summaries will be reduced.
But this approach is reactive. Users can only delete signals that are considered irrelevant, not actively telling Spotify about their new preferences. This is where the Notes feature is allegedly going to take a role.
Based on the code snippets found, users will later be able to add written notes to "influence what you see on Home." One string reads, "Tell us more about you," while another states, "Your notes help influence what you see on Home." There are even examples of text placeholders such as, "I've been listening to a lot of...", which indicates that Spotify is opening up a free input space for users.
Conceptually, this is an interesting shift. Instead of relying entirely on behavior-based machine learning, Spotify seems to want to combine explicit signals from users. In simple terms: not just guessed, but also told directly.
However, there are limits. From the same code it is revealed that the number of records that can be created will be limited, including the number of characters per record. If the limit is reached, users must delete old records before adding new ones.
The removal of the note is also said to "reduce its impact" on Taste Profile, indicating that the system will take into account the historical weight of each input.
Android Authority reported that this feature has not yet been activated, indicating that it is still in the development stage and is not yet available to the public. Spotify itself has not given an official statement regarding the launch schedule or the detailed mechanism.
In addition to the Notes feature, code deconstruction also reveals potential updates to Spotify Messages. A string was found that read, "Pick custom emoji reactions for chat messages," which hinted at plans to introduce custom emoji reactions in conversations.
Currently, Spotify Messages only provides six standard emojis for reactions. If custom emoji support is actually launched, users will have more flexibility in responding to messages - a small step that could strengthen the social aspect of the application.
Spotify's move shows a wider trend in the tech industry: algorithms are getting more sophisticated, but users also want more control. In an era of extreme personalization, people are no longer content to just be a data point. They want to get in on the "driving" the system that recommends their daily entertainment.
If the official Notes feature is released, Spotify is not just a streaming platform with smart AI, but also a space for dialogue between humans and algorithms. And in a world that is increasingly curated by machines, the ability to say, "Hey, this is my taste now," can feel revolutionary.
The English, Chinese, Japanese, Arabic, and French versions are automatically generated by the AI. So there may still be inaccuracies in translating, please always see Indonesian as our main language. (system supported by DigitalSiber.id)
