READ ME File For 'Dataset: Code for Psychoacoustic Metrics used in "Practical Audio System Design for Private Speech Reproduction"' Dataset DOI: 10.5258/SOTON/D1486 ReadMe Author: Daniel Wallace, University of Southampton orcid.org/0000-0003-0212-5395 This dataset supports the PhD Thesis entitled "Practical Audio System Design for Private Speech Reproduction" by Daniel Wallace This dataset contains: Metrics.zip - A zip archive containing: Annoyance.m - MATLAB function to calculate the Psychoacoustic Annoyance of a given signal. FluctuationStrength.m - MATLAB function to calculate the Fluctuation Strength of a given signal. Sharpness.m - MATLAB function to calculate the Sharpness of a given signal. Noise-30s.wav - File containing speech-shaped noise with same spectrum as Speech-30s.wav setSPL.m - MATLAB function to set the representative Sound Pressure Level of a given signal Speech-30s.wav - File containing recorded male speech. Recordings are derived from the Hurricane Natural Speech Corpus (https://doi.org/10.7488/ds/2482) which is licensed under a CC BY-NC 4.0 License. SPL.m - MATLAB function to calculate the representative Sound Pressure Level of a given signal test_Annoyance.m - MATLAB script to test each of the metrics in this dataset, using the supplied .wav files. Roughness.m - MATLAB function to calculate the Roughness of a given signal. This code is derived from the PsySound3 Project (http://psysound.wikidot.com), which is licensed under a CC BY-SA 3.0 License readme.txt - This file Dependencies: Code in this dataset requires the Genesis Loudness Toolbox to run. This can be downloaded from the Internet Archive: https://web.archive.org/web/20190509124114/http://genesis-acoustics.com/en/pages/post/sonie.php (Accessed 29/07/2020) MATLAB is required to run the *.m files in this dataset. Dataset was created and tested using MATLAB R2019a, but may run on other versions of MATLAB. Licence Information: Metrics.zip and readme.txt: CC BY-NC 4.0 Roughness.m: CC BY-SA 3.0 Related projects: EPSRC Centre for Doctoral Training in Next Generation Computational Modelling - Grant EP/L015382/1 Date that the file was created: 29/07/2020