init_Wav2Vec

This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6622
• Accuracy: 0.6757

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.73	2	0.6950	0.4865
No log	1.82	5	0.6838	0.4865
No log	2.91	8	0.6538	0.7027
0.6755	4.0	11	0.6984	0.4865
0.6755	4.73	13	0.7010	0.4054
0.6755	5.82	16	0.6609	0.7027
0.6755	6.91	19	0.6618	0.6757
0.6234	7.27	20	0.6622	0.6757

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.2
• Datasets 2.16.1
• Tokenizers 0.13.3