The Short-Time Objective Intelligibility measure
d = stoi(x, y, fs_signal) returns the output of the short-time
objective intelligibility (STOI) measure described in [1, 2], where x
and y denote the clean and processed speech, respectively, with sample
rate fs_signal in Hz. The output d is expected to have a monotonic
relation with the subjective speech-intelligibility, where a higher d
denotes better intelligible speech. See [1, 2] for more details.
References:
[1] C.H.Taal, R.C.Hendriks, R.Heusdens, J.Jensen 'A Short-Time
Objective Intelligibility Measure for Time-Frequency Weighted Noisy
Speech', ICASSP 2010, Texas, Dallas.
[2] C.H.Taal, R.C.Hendriks, R.Heusdens, J.Jensen 'An Algorithm for
Intelligibility Prediction of Time-Frequency Weighted Noisy Speech',
IEEE Transactions on Audio, Speech, and Language Processing, 2011.
Copyright 2009: Delft University of Technology, Signal & Information
Processing Lab. The software is free for non-commercial use. This program
comes WITHOUT ANY WARRANTY.
Updates:
2011-04-26 Using the more efficient 'taa_corr' instead of 'corr'