Song Scope frog identification failed on roughly half of recordings

Evidence

For instance, its application to study the long-term dynamics of frogs has been tried by Waddle et al. (2009), processing 200 h of recordings of three species of frogs: green treefrog (Hyla cinerea), American bullfrog (Lithobates catesbeianus), and bronze frog (Lithobates clamitans). Identification of the different species of frogs was biased by false positives (2.7–15.8 % per species) and failed to identify species (false negatives) in 45–51 % of recordings. … This method requires a prudent approach to be used for a large monitoring network. — Almo Farina, Soundscape Ecology, 2055

Categorization

• recurring: this claim recurs 2 times within its source
• XREF: connects to: Automated vocalization identification still fails under masking from geophony and anthrophony — Almo Farina, Soundscape Ecology: Principles,…, 2049; Automated Gabor-transform identification succeeded on 67 bird species but struggled with low-frequency-dominated songs — Almo Farina, Soundscape Ecology: Principles,…, 2054; Wind and rustling leaves are a major obstacle to recording biophonies — Almo Farina, Soundscape Ecology: Principles,…, 292

Broader Connections

• Automated Gabor-transform identification succeeded on 67 bird species but struggled with low-frequency-dominated songs #corroborates (0.85) — Almo Farina, Soundscape Ecology, 2054
• Wind and rustling leaves are a major obstacle to recording biophonies #qualifies (0.72) — Almo Farina, Soundscape Ecology, 292