wav2vec2-large-xlsr-53-Hindi-Version1

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

• Loss: 0.7287
• Wer: 0.5457

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: 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: 500
• num_epochs: 70
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.6017	6.7568	500	3.5280	1.0
3.3879	13.5135	1000	3.3755	1.0
3.3566	20.2703	1500	3.3544	1.0
3.3133	27.0270	2000	3.2753	1.0
2.216	33.7838	2500	1.8757	0.9159
1.2972	40.5405	3000	1.0386	0.6969
1.0939	47.2973	3500	0.8590	0.6190
1.0188	54.0541	4000	0.7791	0.5797
0.9468	60.8108	4500	0.7461	0.5575
0.9806	67.5676	5000	0.7287	0.5457

Framework versions

• Transformers 4.45.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1