[christiansafka]

We've published our learnings and results as part of our entry to the Interspeech 2026 Audio Encoder challenge. Code and model will be released on the same url