wav2vec2-base-timit-finetune-4-additonal-train

This model is a fine-tuned version of rohitp1/wav2vec2-base-timit-finetune-3 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4123
• Wer: 0.2905

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 60
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1543	2.77	200	0.3600	0.3153
0.1277	5.55	400	0.3621	0.3110
0.106	8.33	600	0.3862	0.3050
0.095	11.11	800	0.4055	0.3042
0.0862	13.88	1000	0.4001	0.3083
0.0834	16.66	1200	0.4228	0.2993
0.0676	19.44	1400	0.4245	0.2955
0.0593	22.22	1600	0.4123	0.2905

Framework versions

• Transformers 4.24.0
• Pytorch 1.12.1
• Datasets 2.7.0
• Tokenizers 0.11.0