wav2vec2-xls-r-asheshi-akan-5-hours

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3649
• Wer: 0.2890
• Cer: 0.1532

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: 32
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
6.9205	5.6818	500	3.1754	1.0	1.0
3.0873	11.3636	1000	2.8653	1.0	1.0
1.4627	17.0455	1500	1.0507	0.4057	0.2008
0.4223	22.7273	2000	1.0953	0.3260	0.1708
0.2799	28.4091	2500	1.1691	0.3063	0.1633
0.2197	34.0909	3000	1.3488	0.3014	0.1655
0.1707	39.7727	3500	1.3593	0.2932	0.1566
0.1454	45.4545	4000	1.3649	0.2890	0.1532

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0