Task
Automatic Speech Recognition
W
sumedh/wav2vec2-large-xlsr-marathi
Automatic Speech Recognition·sumedh· 642.7K· 2
transformers
apache-2.0
315.5M params
dataset:openslrbase_model:facebook/wav2vec2-large-xlsr-53base_model:finetune:facebook/wav2vec2-large-xlsr-53
Wav2Vec2-Large-XLSR-53-Marathi Fine-tuned facebook/wav2vec2-large-xlsr-53 on Marathi using the Open SLR64 dataset. When using this model, make sure that your speech input is sampled at 16kHz. This data contains only female voices but the model works well for male voices too. Trained on Google Colab Pro on Tesla P100 16GB GPU. WER (Word Error Rate) on the Test Set: 12.70 % Usage The model can be u
# pull & run locally
pip install mlforge-sdk && mlforge pull sumedh/wav2vec2-large-xlsr-marathi
pip install mlforge-sdk && mlforge pull sumedh/wav2vec2-large-xlsr-marathi
Model details
Provider
sumedh
Framework
transformers
Parameters
315.5M
Size
2.4 GB
License
apache-2.0
Downloads
642.7K
Likes
2
Updated
2025-02-22
About sumedh/wav2vec2-large-xlsr-marathi
Wav2Vec2-Large-XLSR-53-Marathi Fine-tuned facebook/wav2vec2-large-xlsr-53 on Marathi using the Open SLR64 dataset. When using this model, make sure that your speech input is sampled at 16kHz. This data contains only female voices but the model works well for male voices too. Trained on Google Colab Pro on Tesla P100 16GB GPU. WER (Word Error Rate) on the Test Set: 12.70 % Usage The model can be used directly without a language model as follows, given that your dataset has Marathi actualtext and pathinfolder columns: