Time Scale Modification Subjective Quality Assessment

Greetings to WASPAA and SANE conference attendees.
If you are looking for the MATLAB implementations you can find them at github.
If you are looking for the Time-Scale Modification Database, you can find the unlabelled data here.
The labelled data will be released once the accompanying paper has been published.

Initial subjective testing has now finished, however additional sets may be availble in the future.
It should take approximately 15 minutes to complete each set. The sliders start in a random position and have no connection to the ratings given already.
And as a reminder, please use headphones or high quality speakers when rating the audio files.

Before starting, I would suggest reading through this tutorial, which will explain the process and give examples of what you are listening for.
Please read through the Information and Consent forms that can be found here. At the start of the testing you'll be asked if you have read them and give consent for doing the testing.
A big thank you to Nicholas, Brecht and David for their Web Audio Evaluation Tool that has been used to setup this testing. More information can be found here.
And finally, thank you in advance for being involved with this subjective testing.
Please note that this testing makes use of HTML5 for audio playback. I have tested successfully with Chrome and Firefox on Windows and Mac, as well as iOS.

The results so far look like the plot below. The red line is the Mean Opinion Score for each file. Ratings go above and below the rankings due to normalisation of ratings.


Griffith University ethics approval: 2018/671
Last Updated: 6/2/20 1700AEST