Every dog owner knows the moment. Your dog’s breathing sounds different tonight: raspier, faster, a snore that wasn’t there last month. Is it serious, or nothing? Today, nobody can actually measure it. You describe it to your vet, maybe show a video, but breathing changes minute to minute and a single check-up rarely catches what you heard at home.
We are changing that. Fauna Acoustics, turns any smartphone into an instrument that measures a dog’s breathing. Hold the phone near your dog, record for a few seconds, and Fauna produces a clear, numbers-based summary your vet can act on. No extra hardware, nothing for the dog to wear.
Why breathing
Breathing is one of the earliest and most honest signals of a dog’s health. Changes in breathing rate and sound often come before other signs of heart, airway and lung trouble, and flat-faced breeds like Pugs and French Bulldogs — among the world’s most popular — live with airway problems their whole lives. Yet in the clinic, breath sounds are still judged by ear alone. The WSAVA position paper on lung auscultation acknowledges that finer grading of breath sounds was set aside because no instrument existed to do it.
Digital listening is already established in human medicine. In veterinary care, no one yet measures breathing sounds in animals. Fauna is building that breathing layer, designed for animals from day one.
How it works
Fauna combines custom signal processing with machine learning in three stages.
1. Separating the dog from the room. A home recording is full of fans, traffic, voices and footsteps. Fauna’s source-separation model, based on a technique called non-negative matrix factorisation, splits the recording into its component sounds and keeps only the breathing. A dedicated detector recognises footsteps and bumps so they are never counted as breaths.
2. Measuring every breath. At Fauna’s core is a custom digital signal processing (DSP) engine, deliberately built not to be a black box. It finds and times each breath cycle, then measures what a vet listens for: rate and rhythm, how much of a sound is clear tone versus turbulent noise (harmonic-to-noise ratio), how harsh it is, and where in the airway it most likely originates. Every number is stored with a record of how it was calculated, the evidence used, and how confident Fauna is.
3. Putting it in context. A machine-learning layer is trained to recognise nine types of sound, including normal breaths, coughs, snoring and wheeze. Results are compared with what is normal for the dog’s breed, age, size and breathing speed, and with its own earlier recordings on the same phone, so each microphone’s quirks cancel out. It works on the phone you already own.
If a recording is too noisy or unclear, Fauna says so and shows no number at all. Every result ends with “For veterinary evaluation.” Fauna measures; the vet decides.
Where Fauna is today
In a matter of months, Fauna has gone from idea to working prototype: its own iPhone app with built-in recording guidance, breathing rate calculated on the phone, a vet dashboard, a pet-parent screen, and a secure connection for vet platforms to send recordings and receive results.
Early testing is strong. Fauna measured breathing rate with high accuracy in pilot test cases, from slow rest to fast panting. Across 127 noise-only recordings, it never once reported breathing that wasn’t there. It processed the real recordings cleanly, and more than 250 automated checks pass.
Fauna is also following individual dogs across repeated sessions to build a personal baseline for each animal. This tests what matters most for long-term monitoring: whether Fauna can spot when a dog’s breathing moves away from its own normal — and confirms it is learning the dog, not the room.
Jatin Sharma – Founder, Fauna Acoustics
” Our aim is to make veterinary technology accessible, affordable, and reliable. We’re starting with the breath, because it’s the most basic sign of life and one of the hardest to measure well. From there, we’ll build tools that help catch problems earlier, follow each animal’s health over time, and give animals longer, better lives “
The opportunity
Fauna is built as a platform, not a single app. The same engine can plug into vet apps, telemedicine services and pet insurers on a per-analysis basis, and be licensed into smart collars and veterinary stethoscopes. The roadmap runs from clinic validation with vet partners, to platform partnerships to hardware integrations, and eventually Fauna’s own wearable.
Its edge rests on three things that are hard to copy: measurement adjusted for breed and breathing speed, independence from any particular device, and results that can be checked step by step — the transparency vets, insurers and regulators look for.
The founder
Fauna is founded by Yatin Sharma, who spent a decade in IT infrastructure, security and VoIP (internet voice) systems — work centred on carrying sound faithfully and building systems people trust. Fauna is built in Mohali, India, for vets and animal-health partners worldwide, under Bodh Informatics, a brand of DPIIT-recognised SHYL Collective Private Limited. Fauna is now inviting veterinary clinics to join early validation, and partners across vet technology and animal health to build with it. Reach Yatin at yatin@faunacoustics.com or visit faunacoustics.com.
About The Author
