K. S. Ernest (iFire) Lee

The Godot-whisper project has been on pause because we have had poor experiences with the real-time performance. Been focusing on the core part of V-Sekai rather than the speech-to-text model.

> https://godotengine.org/asset-library/asset/2638 > > Hello everyone. Please tell me, I want to launch and use Godot Whisper, but after adding the node Speech To Text I don't have a section...

Can you make sure that your meshes are skinned? Like each vertex is associated with a skeleton.

Discussed the usefulness of marketing the need to champion gltf extension rather than merely submitting specs.

I was interested in getting a qwen3 vision model working like https://huggingface.co/huihui-ai/Huihui-MiniCPM-V-4_5-abliterated

Godot Engine has two discord communities. The official discord ran by the Godot Engine Foundation and Godot Cafe. https://godotengine.org/community/

In gltf document we have a gltf extension system that converts this. But I was waiting for demand for this feature to restore it. I’ll post a gdscript version momentarily...

https://github.com/V-Sekai/TOOL_model_explorer/blob/cb9d112f051ba5426cd04b6e95228f03c30c744f/vsk_model_explorer/core/UIController.gd#L85 Note that this has a bug where I free empty nested importer mesh instances in error

My original plan was to add a FBXDocument extension system and then use FBX runtime import as the conversion of ImporterMesh3D to MeshInstance3D as a runtime plugin. I think that...

That was the original design, but when we tried that the gltfdocument and fbxdocument didn't gain much from separation. We also had to duplicate the 10-20 auxiliary classes like `Ref`.