Context is not state: how adaptive sound works
Most adaptive sound works from context: inputs such as the time of day, the weather, your location and, on some devices, your heart rate, which reshape the sound as they change. Context is useful, but it is not state, and the same Tuesday afternoon can find you focused, flat or frazzled.
R13 Labs builds sound that starts from the state instead, and works toward the one you need.
How does adaptive sound work today?
Adaptive sound is sound that changes in response to inputs. Most adaptive sound follows circumstances such as time, weather or location.
Endel, for example, says its technology creates soundscapes that adapt in real time to inputs like time of day, weather, heart rate and location (Endel (opens in a new tab)). Those inputs are well chosen: they are easy to measure, and they say a lot about the situation you are in.
Mostly, they describe the situation. How you are within it is a different question.
Why isn't context enough?
Because the same situation finds different people in different states, and the same sound lands differently on each of them. Research on background music shows this plainly.
A 2011 meta-analysis found no overall effect of background music across studies, and its authors concluded that the null result most likely came from positive and negative effects cancelling out (Kämpfe, Sedlmeier and Renkewitz (opens in a new tab)). An average that hides real effects is a warning to any system that gives everyone the same answer.
Who is listening matters. In a study of 76 people, background music and office noise lowered scores on reading, recall and arithmetic compared with silence, and introverts were hit harder than extraverts (Furnham and Strbac, 2002 (opens in a new tab)).
What is playing matters too. In another experiment, instrumental music was most likely to disrupt reading comprehension when it was both fast and loud (Thompson, Schellenberg and Letnic, 2012 (opens in a new tab)).
Psychologists explain much of this through arousal and mood. Sound shifts both, and whether that helps depends on the person, the task and where they started (Thompson, Schellenberg and Husain, 2001 (opens in a new tab); Furnham and Strbac, 2002). A rule that maps the hour or the weather to a sound gives everyone in that hour the same answer.
What do we mean by state?
By state we mean how a person is doing right now: focused, flat or frazzled. It is not a mood label you pick, and it is not the time of day.
Our engine forms a picture of state from the signals a product gives it, along with a sense of how sure it is. In a study session in Aqademiq, those signals include how the session is going and what you tell it.
From that picture it sets a direction, toward focus, calm or energy, and decides how far to go. R13 Labs calls this cognitive state intelligence: a live read of human state, and sound and haptics generated to meet it.
How does sound follow state without chasing noise?
By holding a target and moving gradually. The engine holds a target state, works on the gap and re-reads state continuously, moving gradually so nothing jumps.
And by knowing how far to trust its own read. A system that reacts to every signal chases noise: the sound swells, drops and swells again, and you notice.
Our engine weighs every move by how sure it is. Unsure, it holds. Sure, it commits.
| Approach | What it starts from | What it does with it |
|---|---|---|
| A playlist | Tracks chosen in advance | Plays what was chosen, whatever is happening |
| Context-led adaptive sound | The situation: time of day, weather, location, sometimes heart rate | Reshapes the sound as those inputs change |
| State-led sound, as we build it | An estimate of the person's state, and how sure that estimate is | Holds a target, works on the gap, and moves only as far as it is sure |
The technology page explains the engine at the level of what it does.
Does a room have a state?
Partly. Some parts of state are meaningful across a group. Some belong to an individual and don't survive being averaged.
That is why Prism Venues, our product for spaces, works differently from a study session today. It gives each zone of a space an original sound and shapes it across the trading day toward what that zone is set to feel. It has no microphones and records no one.
Staff can lift or settle the room with Mood Nudge, or take full control with Takeover. Prism Venues doesn't read the room's state; it holds the one each zone was designed for.
What don't we know yet?
The engine estimates, and it is sometimes wrong. That is why how sure it is shapes how far it moves.
We test two questions, because they can fail separately: does the read match reality, and does the sound move state the way it intends to? Work is under way on the first, in the study context, and we'll publish what it finds.
Whether our engine does this better than a good playlist is not settled yet. Why we won't claim more than the evidence supports, starting with brainwave entrainment, is the subject of Do binaural beats work?
Frequently asked questions
What is adaptive sound?
Adaptive sound is sound that changes in response to inputs. Most adaptive sound follows circumstances such as time, weather or location, and some also uses heart rate. R13 Labs builds sound that starts from an estimate of a person's state instead, and moves toward the state they need.
What is the difference between context and state?
Context is the situation around you: the hour, the weather, where you are. State is how you are within it: focused, flat or frazzled. The same context can hold very different states, which is why R13 Labs builds sound that starts from state.
What is cognitive state intelligence?
Cognitive state intelligence is R13 Labs' term for a live read of human state, and sound and haptics generated to meet it. Its engine holds a target, works on the gap, and moves only as far as it is sure of its read.
Does Prism Venues listen to guests?
No. Prism Venues has no microphones and records no one. It gives each zone of a space an original sound, shaped across the trading day toward what that zone is set to feel, and staff can nudge it or take it over.
Does R13 Labs claim its sound improves focus?
Not yet. R13 Labs builds on the established finding that sound shapes attention, arousal and mood, and tests its own engine against two questions, starting with whether its read matches reality.
R13 Labs builds a real-time engine that shapes sound to the state a person or a space needs. The engine runs Aqademiq, a focus app on iOS and Android, and Prism Venues, original sound for every zone of hotels, cafes, clinics, stores and offices in India and the UAE. R13 Labs is based in Abu Dhabi and Coimbatore.
Read the science.
What we've established, what we're testing and what we don't claim, on our science page.


