This model is a fine-tuned version of facebook/wav2vec2-base-960h on the audiofolder dataset. It achieves the following results on the evaluation set:

Loss: nan
Wer: 1.0

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: 0.0001
train_batch_size: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 4
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 5
num_epochs: 50
mixed_precision_training: Native AMP

Training results

Epoch	Step	Validation Loss	Wer
2.0	10	nan	1.0
4.0	20	nan	1.0
6.0	30	nan	1.0
8.0	40	nan	1.0
10.0	50	nan	1.0
12.0	60	nan	1.0
14.0	70	nan	1.0
16.0	80	nan	1.0
18.0	90	nan	1.0
20.0	100	nan	1.0
22.0	110	nan	1.0
24.0	120	nan	1.0
26.0	130	nan	1.0
28.0	140	nan	1.0
30.0	150	nan	1.0
32.0	160	nan	1.0
34.0	170	nan	1.0
36.0	180	nan	1.0
38.0	190	nan	1.0
40.0	200	nan	1.0
42.0	210	nan	1.0
44.0	220	nan	1.0
46.0	230	nan	1.0
48.0	240	nan	1.0
50.0	250	nan	1.0

Framework versions

Transformers 4.57.3
Pytorch 2.9.1+cpu
Datasets 3.6.0
Tokenizers 0.22.1

Muayad2025/wav2vec2-base-960h-kurdish-custom

wav2vec2-base-960h-kurdish-custom

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Muayad2025/
wav2vec2-base-960h-kurdish-custom