Divide and Remaster v3: Indo-Aryan Test Set

<p>Divide and Remaster v3 is a multilingual rework of the Divide and Remaster v2 dataset by P&eacute;termann et al.</p> <p>This repository contains the&nbsp;<strong>Indo-Aryan variant </strong>of DnR v3.</p> <p>The major changes from DnR v2 are as follows:</p> <ul> <li>the dialogue stem now contains content from more than 30 languages across various language families;</li> <li>speech, vocals, and/or vocalizations have been removed from the music and effects stems;</li> <li>loudness and timing parametrization have been adjusted to approximate the distributions of real cinematic content;</li> <li>the mastering process now preserves relative loudness between stems and approximates standard industry practices.</li> </ul> <p>See the linked GitHub repository for more details.</p>