metadata
language:
- 'no'
- nb
Warmstarted from Chills model, then trained for 25 (de facto 50) epochs. Batch size 16, learning rate (√2)e-3 for the first 15(?) epochs and (5√2)e-4 for the last 10.
Dataset: NST Norwegian Speech Synthesis (CC0), shuffled together with a copy that has had all audio files under 6(?) seconds merged recursively.