w2v2-libri-10min – AI Model by lauramartinezes | AlphaNeural AI

You can deploy this model and start earning money today!

lauramartinezes/
w2v2-libri-10min

like

0

automatic-speech-recognition

generated_from_trainer

endpoints_compatible

Views

No views yet

w2v2-libri-10min

This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 2.1030
Wer: 0.6113

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0003
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
training_steps: 2500
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
5.1068	62.5	250	2.9012	1.0
0.6528	125.0	500	1.5705	0.6404
0.0697	187.5	750	1.7490	0.6570
0.0352	250.0	1000	2.0035	0.6252
0.033	312.5	1250	2.1108	0.6432
0.0176	375.0	1500	2.0904	0.6473
0.0072	437.5	1750	2.1286	0.6127
0.0048	500.0	2000	2.1334	0.6279
0.0043	562.5	2250	2.0559	0.6100
0.0027	625.0	2500	2.1030	0.6113

Framework versions

Transformers 4.29.2
Pytorch 2.0.1+cu118
Datasets 1.18.3
Tokenizers 0.13.3