Long-duration field audio recording creates a simple but stubborn problem. Data adds up fast. High sample rates, long deployments, and continuous recording can overwhelm storage, especially as microSD cards become more expensive and harder to source.
To address this, Owl Sense supports WavPack lossless audio compression directly on the device. This post walks through why we chose WavPack, how it is configured on Owl Sense, and how to convert recordings back to standard WAV for analysis.
Why WavPack?
WavPack is a lossless audio compression format designed specifically for high-quality audio.
For bioacoustics and environmental monitoring, this matters because:
- No audio data is lost through bit-for-bit reconstruction
- Recordings remain suitable for spectral analysis, detection, and machine learning models
- Compression happens during recording, not as a post-processing step
In real-world field conditions, Owl Sense recordings typically see 50 to 80 percent size reduction, depending on ambient noise and signal complexity.
WavPack for Owl Sense: Built In
WavPack support is native to the current Owl Sense firmware and app.
- Audio is recorded directly to microSD in WavPack format
-
Compression occurs on-device in real time
From the user’s perspective, recording with WavPack feels exactly like recording WAV, except storage lasts much longer.
Configuring WavPack on the Device
WavPack is enabled through the Owl Sense app as part of your recording configuration. Owl Sense supports two WavPack Audio Format Options:
- WavPack Lossless(.WV LL)
- WavPack Lossy(.WV LY)
We recommend Lossless Audio for almost all use cases. Lossy sounds identical to me a human but may throw away subtle detail that matters to Machine Learning Models and other analysis. Lossy will increase runtimes and might be an option for certain applications but test before hand.
To configure WavPack, set Audio Format to WavPack Lossless (.WV LL in the app)
Once selected, the device handles everything automatically. There is no need to manually manage files, compression levels, or post-recording processing in the field.
What Gets Stored on the SD Card
On the microSD card, recordings are stored as .wv files using the WavPack format.
-
Fully lossless audio
- Safe for long-term archival
- Smaller size than equivalent WAV files
This reduced footprint allows users to:
- Use smaller capacity cards
- Extend deployment duration
- Avoid relying on high capacity cards that may be expensive or unavailable
Converting WavPack Back to WAV
WavPack files can be converted back to standard WAV at any time with no loss of quality.
Option 1: Command-Line Conversion
For advanced users or automated pipelines, WavPack provides a lightweight command-line tool.
Example:
wvunpack /audio/*.wv /audio/output
This produces a standard WAV file that is bit-for-bit identical to what would have been recorded without compression.
These WAV files work seamlessly with:
- BirdNET
- OpenSoundscape
- Raven
- Audacity
- Custom machine learning pipelines
Option 1: Paid UI tool, dBpoweramp
dBPoweramp's Music Converter tool supports converting from WavPack to almost any audio format you could imagine. It's easy to use and fast. Cost is a one time $48-$85 depending on personal or business use.
https://www.dbpoweramp.com/dmc
Does Compression Affect Analysis?
No, not with lossless audio.
Because WavPack is lossless, all downstream analysis sees the exact same audio data as an uncompressed WAV file. Frequency content, amplitude, timing, and noise characteristics are preserved.
In practice, this means:
- No impact on detections
- No changes to spectrograms
- No need to retrain models
The only difference is storage efficiency.
Why This Matters Right Now
Global demand for NAND flash, driven largely by AI infrastructure, has increased the cost and volatility of microSD card supply. By reducing storage requirements at the source, Owl Sense users are less exposed to:
- Price spikes
- Availability shortages
- Forced upgrades to larger cards
Compression provides flexibility without compromise.
Designed for the Field
Our goal with WavPack support was not just to save space. It was to make long-term field recordings run longer, more resilient, affordable, and predictable.
With Owl Sense:
- WavPack is already supported
- Compression is automatic
- Audio integrity is preserved
- Existing analysis workflows remain unchanged
Smarter storage means fewer constraints and more time focused on the science.
Owl Sense
Happy recordings 🦉🎙️