metadata

license: mit
base_model: facebook/w2v-bert-2.0
tags:
  - generated_from_trainer
datasets:
  - common_voice_17_0
metrics:
  - wer
model-index:
  - name: w2v-bert-2.0-armenian-CV17.0
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_17_0
          type: common_voice_17_0
          config: hy-AM
          split: test
          args: hy-AM
        metrics:
          - name: Wer
            type: wer
            value: 0.12880886426592797

w2v-bert-2.0-armenian-CV17.0

This model is a fine-tuned version of facebook/w2v-bert-2.0 on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:

Loss: 0.1202
Wer: 0.1288
Cer: 0.0227

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: 5e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
1.6647	1.0	325	0.2207	0.2605	0.0452
0.1807	2.0	650	0.1774	0.2183	0.0382
0.111	3.0	975	0.1447	0.1671	0.0295
0.0672	4.0	1300	0.1303	0.1439	0.0252
0.04	5.0	1625	0.1202	0.1288	0.0227

Framework versions

Transformers 4.40.1
Pytorch 2.2.1+cu121
Datasets 2.19.0
Tokenizers 0.19.1