Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Eval Results (legacy)
Instructions to use misterkissi/w2v2-lg-xls-r-300m-deg with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use misterkissi/w2v2-lg-xls-r-300m-deg with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="misterkissi/w2v2-lg-xls-r-300m-deg")# Load model directly from transformers import AutoProcessor, AutoModelForCTC processor = AutoProcessor.from_pretrained("misterkissi/w2v2-lg-xls-r-300m-deg") model = AutoModelForCTC.from_pretrained("misterkissi/w2v2-lg-xls-r-300m-deg", device_map="auto") - Notebooks
- Google Colab
- Kaggle
w2v2-lg-xls-r-300m-deg
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m.
It achieves the following results on the evaluation set:
- eval_wer: 0.09674
- eval_runtime: 185.7492
- eval_samples_per_second: 33.804
- eval_steps_per_second: 4.226
- epoch: 2.23179
- step: 11400
- Downloads last month
- 4
Model tree for misterkissi/w2v2-lg-xls-r-300m-deg
Base model
facebook/wav2vec2-xls-r-300mSpace using misterkissi/w2v2-lg-xls-r-300m-deg 1
Evaluation results
- WERself-reported9.674