DGR: Gender Recognition of Human Speech Using One-Dimensional Conventional Neural Network

Figure 2

A British English female’s voice sample and its features: (a) voice sample (B_eng_f1.wav); (b) mel-spectrogram; (c) MFCC; (d) chromagram; (e) spectral contrast; (f) tonal centroids (Tonnetz).

(a)

(b)

(c)

(d)

(e)

(f)