Citrinet 1024 Korean
Storyteller installs this model through ghost-story. This repository mirrors version 2.0.0 of citrinet1024-ko in the Storyteller model registry, with a folder per format.
Trained by ypluit/stt_kr_citrinet1024_PublicCallCenter_1000H, licensed CC-BY-4.0.
Converted from the ONNX export OpenVoiceOS/stt_kr_citrinet1024_PublicCallCenter_1000H_onnx at revision 74230b92a06c.
model.safetensorsholds the encoder weights in the layout the MLX loader expects, plus the mel filterbank and window of the NeMo front end.model.onnxis the export with the mel front end spliced in front of the encoder byexport-onnx.py, so it takes the pre-emphasised, reflect padded signal and computes the features and their length itself. The encoder is unchanged.config.jsondescribes the architecture, the front end and the block layout, written byconvert.py.vocab.txtis unchanged from the export.
The onnx format holds model.onnx, the mlx format model.safetensors, each with the configuration, vocabulary and tokenizer.
convert.py and export-onnx.py are in libraries/mlx-audio/scripts.
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support