Animals
Birds and amphibians are two of OneAquaHealth's eleven indicators of stream health, and natural controls of mosquitoes and other disease carriers. Murmur names 11,560 kinds of calls, including 647 amphibians.
A stethoscope for urban streams
Listen to a stream.
Hear its health.
Murmur turns a short recording made at a city stream into a biodiversity, noise and wellbeing reading, confirmed by the people who were there, and sets it beside what the OneAquaHealth labs found in the water.
Free, open source, and private by design: analysis runs in your browser.
Drawing the sound…
One recording, three kinds of health
Environment
Rushing water and traffic leave a measurable mark. Murmur reports the balance of natural and human sound, and how much of the recording was too loud to hear quiet calls, so silence is never read as absence.
People
Hearing birds lifts wellbeing for hours. The OneAquaHealth app already asks how a stream makes people feel; Murmur records what they actually heard.
How it works
- 1
Record anywhere
A phone voice memo, a field recorder, or the 5–10 second video the OneAquaHealth app already asks for. No new hardware.
- 2
Murmur listens on your device
BirdNET runs in the browser, 3 seconds at a time. The recording is never uploaded.
- 3
You confirm by ear
Every call comes with the clip that triggered it. Yes, not sure, or no: the AI suggests, people decide.
- 4
Compare and share
See what the OneAquaHealth labs found at the same site, then export FHIR for health systems and Darwin Core for biodiversity databases.
Ready for OneAquaHealth to adopt
No new fieldwork
The OneAquaHealth app already asks for a 5–10 second video at every assessment. Murmur reads its sound, so every past and future visit can carry a biodiversity reading.
Same sites, same data
Recordings attach to the project's own site codes (C1–C20, BN1–BN20, G1–G22, O1–O20, T1–T25) and sit beside its lab results from the public API.
Same data model
Exports follow the OneAquaHealth FHIR guide (location-oah, observation-indicators-oah, the IG's own birds and amphibians codes) and Darwin Core for GBIF.
Near-zero running cost
The model runs on the visitor's device, so analysis costs nothing on a server. The shared commons fits the free tiers of Vercel and Neon.
Built on OneAquaHealth, not beside it
The OneAquaHealth citizen app already asks people to describe a stream and say how it made them feel, and its own research found that people keep contributing when they can see what their observations mean. Murmur adds a listening step to that journey: a biological indicator taken from sound, an answer on the spot, and a direct link to the project's research sites, lab results and FHIR data model.
How we checked it, and what it cannot do- OneAquaHealth cities
- 5
- research sites linked
- 106
- species and sound classes
- 11,560
- of audio uploaded to analyse
- 0 bytes