Extended Data Fig. 7: Schematic detailing the three separate pipelines implemented to evaluate ABCS. | Nature Machine Intelligence

Extended Data Fig. 7: Schematic detailing the three separate pipelines implemented to evaluate ABCS.

From: Audio-based AI classifiers show no evidence of improved COVID-19 screening over simple symptoms checkers

Extended Data Fig. 7: Schematic detailing the three separate pipelines implemented to evaluate ABCS.The alternative text for this image may have been generated using AI.

openSMILE- SVM baseline, the Bayesian Neural Network (BNN) and the Self-Supervised Audio Spectrogram Transformer (SSAST). Both SSAST and BNN first convert the raw audio signal to mel spectrogram space whereas the openSMILE–SVM approach extracts a series of handcrafted audio features.

Back to article page