wav2vec2-large-xlsr-53-AsanteTwi-06second

This model is a fine-tuned version of ayertey01/wav2vec2-large-xlsr-53-AsanteTwi-06 on the common_voice_13_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.6678
• Wer: 0.4375

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: 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: 200
• num_epochs: 120

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.0992	16.67	100	2.4101	1.0
0.3622	33.33	200	0.8652	0.4375
0.1804	50.0	300	0.8567	0.4375
0.1412	66.67	400	0.8914	0.5625
0.1193	83.33	500	0.6723	0.4375
0.1063	100.0	600	0.6634	0.4375
0.0976	116.67	700	0.6678	0.4375

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.13.1
• Tokenizers 0.13.3