How Should We Extract Discrete Audio Tokens from Self-Supervised Models? Paper • 2406.10735 • Published Jun 15, 2024