





New sounds added!
The field guide is found on the "Classify" page, in a white box on the right hand side of the screen.
The purpose of Forest Sounds is to study the biodiversity and human threats to tropical forests using sound recordings from acoustic sensors. We aim to use this data to inform conservation measures and help protect the forest.
Our recordings are obtained from 10 AudioMoth acoustic sensors that we have deployed in the Tapir Mountain Nature Reserve, a tropical forest reserve in central Belize. The data were recorded between April 2018 to 2019.
We will analyse information on threats such as hunting and logging to better understand where and when these threats occur. We will share this information with rangers in Belize to help them better protect the forest. For example, by sharing information on where and when hunting occurs, they can target these areas and times for ranger patrols. Information on biodiversity will be used to better understand the animal species occurring in Tapir Mountain, as little is known about the biodiversity occurring in this remote area.
If you can't identify a sound, go to our Field Guide to listen to examples of sounds commonly recorded, and then listen to the recording again. It is also very informative to look at the image whilst listening to the sound, and then compare this image with the examples in the Field Guide. If you're still unsure, you can flag up any subject you don't feel sure about by clicking on the Talk button at the end of your classification. You can leave a comment or question and we will get back to you.
Join us on Talk and we will get back to you as soon as possible.
Please select all of the labels that correspond to the sounds you hear.
If the sound is not listed among the answers then select 'None of the above'. If there is something interesting about the sound please flag it up by clicking on the Talk button at the end of your classification.
This image is called a spectrogram. This is a way of visualising the strength or 'loudness' of the signal over time at different frequencies.
Sometimes our sensors record human speech, although this is nearly always the sounds of our own voices as we are setting up the sensors.
No - we will not collect any additional information to identify the person speaking (if we happen to record someone not on our research team). As the clips are only 4 seconds long, any conversations would be sufficiently interrupted to keep them private.