A text-to-speech model trained on norwegian parliament datasets (2000h) for 3 epochs. Based on Qwen/Qwen2.5-0.5B and outputs wavtokenizer tokens.

Downloads last month
2
Safetensors
Model size
498M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.