Skip to main content
. 2017 Jun 23;141(6):4705–4714. doi: 10.1121/1.4986931

FIG. 8.

FIG. 8.

Comparison of speaker-independent models (trained on 77 speakers and tested on six unseen speakers) and speaker-dependent models in terms of STOI. Group means and standard errors are shown. (a) Performance with the unseen babble noise. (b) Performance with the unseen cafeteria noise.