Instructions to use fidoriel/moonshine-tiny-de with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use fidoriel/moonshine-tiny-de with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="fidoriel/moonshine-tiny-de")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("fidoriel/moonshine-tiny-de") model = AutoModelForSpeechSeq2Seq.from_pretrained("fidoriel/moonshine-tiny-de") - Notebooks
- Google Colab
- Kaggle
Moonshine
This is a, to German, transfer learned moonshine tiny based on the English UsefulSensors/moonshine-tiny The new weights in this repository are licensed under Attribution-NonCommercial-ShareAlike 4.0 International.
Original Inference Code: https://github.com/moonshine-ai/moonshine ONNX Rust Port: https://github.com/fidoriel/moonshine_onnx_rs
Eval
Evaluation on Common Voice test split (https://huggingface.co/datasets/fidoriel/cv-22-de)
- WER: 0.114
- CER: 0.042
Funding Notice
The project on which this model is based was funded by the Federal Ministry of Research, Technology and Space under the funding code “KI-Servicezentrum Berlin-Brandenburg” 01IS22092. Responsibility for the content of this publication remains with the author.
- Downloads last month
- 6