Researchers typically extract distance-related acoustic features from audio, such as the Mel-spectrogram energy envelope based on the short-time Fourier transform, the reverberation decay rate in the cepstral domain, or use deep learning to estimate distance labels directly from speech.