Skip to main content
. Author manuscript; available in PMC: 2016 Dec 1.
Published in final edited form as: IEEE/ACM Trans Audio Speech Lang Process. 2016 Mar 1;24(5):967–977. doi: 10.1109/TASLP.2016.2536478

Fig. 3.

Fig. 3

Comparison of mapping and masking when the number of the utterances of the target speaker is limited. (a) The spectrogram of the utterance of the target speaker. (b) The spectrogram of the first utterance of the interfering speaker. (c) The spectrogram of the second utterance of the interfering speaker. (d) The spectrogram of the mixed signal produced from the target utterance (i.e. Fig. 3a) and the first interfering utterance (i.e. Fig. 3b). (e) The spectrogram of the mixed signal produced from the target utterance and the second interfering utterance (i.e. Fig. 3c). (f) The IRM of the target utterance given the first interfering utterance. (g) The IRM of the target utterance given the second interfering utterance.