Frequently asked questions#
Technical#
What is acoupi?#
acoupi is an open-source Python toolkit that's designed to make it easy to create bioacoustic sensors on edge devices like the Raspberry Pi. acoupi integrates and standardises the entire workflow of bioacoustic monitoring, combining both autonomous recording and classification units. acoupi ensures standardisation of data while providing flexibility of configuration by defining attributes with specific characteristics, the data schema and establishing a set of methods showcasing several behaviours, the components. When configuring acoupi, users configure the set of acoupi components that suit their use case.
What acoupi isn't?#
While acoupi provides modular components to build your own autonomous recording and classification units, it is not a tool for training bioacoustic AI classifiers. acoupi integrates already trained and well-tested AI bioacoustic models through its pre-built programs that you can use to perform on-device bioacoustic classification.
Why acoupi?#
Passive acoustic monitoring (PAM) has emerged as a practical and helpful tool for biodiversity monitoring and conservation. Combining PAM with on-device domain-specific deep-learning bioacoustic classifiers provides opportunities for extending the scale and length of data collection while alleviating downstream data storage and processing burdens. However, deploying and adapting existing solutions for its own use still requires substantial technical expertise.
acoupi aims to provide an all-in-one Python toolkit to make it easy to create your smart bioacoustic sensors on edge devices like the Raspberry Pi. With acoupi, you can either use one of the provided programs or build your custom program using the various tools and components that are included.
What are the available acoupi programs?#
-
default: The acoupi default program is the most simplest program, handling only two tasks: recording and managing audio files.
-
connected: The acoupi connected program extends the default program by adding messaging capabilities, allowing users to send messages to a remote server.
-
acoupi_batdetect2:
-
acoupi_birdnet:
Usage#
Who can use acoupi?#
acoupi is particularly well-suited for researchers in the field of bioacoustics, but it's also a great option for environmental consultants who are looking to set up their own bioacoustic monitoring systems.
Is acoupi free?#
Yes, acoupi is free and is licensed under the GNU GPL-3.0. This means that acoupi will always be free to use, and free to customise for your own purpose. Please do check the license for more detailed information.
I want to use acoupi, but ...#
-
I don't know Python. No problem. While acoupi is built in Python, it includes a variety of pre-built programs that you can use right out of the box. Follow the Tutorials Section to get started.
-
I don't have access to WiFi. acoupi does not require access to WiFi or any other wireless network. You can use acoupi as a standalone device, the audio recordings and audio classification results will be saved on the device's internal storage (i.e., the SD card).
Licensing#
Can I use acoupi for commercial purposes?#
Yes, absolutely! acoupi is licensed under the GNU GPL-3.0, which explicitly permits commercial use. However, the GPL-3.0 is a "copyleft" license. This means that if you modify acoupi or create a new work based on it (whether for commercial or non-commercial purposes), you must make your derived work available under the same GPL-3.0 license. Essentially, this requires you to keep your source code open and freely available.
Does the license apply to the recordings produced by acoupi?#
No, the GPL-3.0 license doesn't apply to the audio recordings themselves. Copyright law governs those. You own the copyright to any recordings you make using acoupi and are free to use them as you wish.
Does the license apply to the detections produced by acoupi?#
This is a bit more nuanced. acoupi itself doesn't provide detection models; it provides the framework for running them. The license of the detection model you use will determine how you can use the detections. If you use a pre-trained model, carefully check its license. Some models may be open source, allowing free use of the detections. Others may have restrictions, especially for commercial use. It's always best to consult with a legal professional for specific advice on licensing and copyright matters, especially when commercial use is involved.
Contributing#
Does acoupi accept outside contributions?#
Yes absolutely, we welcome all contributions: bug reports, bug fixes, documentation improvements, and feature requests.
I want to contribute. Where do I start?#
Amazing! We are always looking for contributors. Make sure to read the contribution guidelines and code of conduct to get started.