Classification Threshold and Training Data Affect the Quality and Utility of Focal Species Data Processed with Automated Audio-Recognition Software

Automated recognition is increasingly used to extract information about species vocalizations from audio recordings. During processing, recognizers calculate the probability of correct classification (“score”) for each vocalization. Our goal was to investigate the implications of recognizer score for ecological research and monitoring. We found a strong relationship between score and the distance at which a call was recorded if the recognizer was trained with calls recorded at close range. These results show that score threshold choice is a decision about sampling area, not just about the balance between false negative and false positive results. Overall, we showed that training recognizers with ‘high-quality’ clips that were recorded at close range will improve the utility of the data without affecting how many true positives the recognizer detects.