mascir_frwav2vec2-large-xlsr-53

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4708
• Wer: 0.3789

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: 8
• 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: 1000
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
6.3318	2.0	500	3.0088	0.9856
1.5916	4.0	1000	0.7746	0.6411
0.3961	6.0	1500	0.5238	0.5211
0.2205	8.0	2000	0.5014	0.4733
0.1401	10.0	2500	0.5166	0.4878
0.1147	12.0	3000	0.5058	0.4333
0.0938	14.0	3500	0.4635	0.4233
0.0788	16.0	4000	0.4997	0.4144
0.0645	18.0	4500	0.4840	0.4122
0.0534	20.0	5000	0.4789	0.4022
0.0437	22.0	5500	0.4785	0.3978
0.041	24.0	6000	0.4708	0.3789

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3