hindi_beekeeping_wav2vec2-3

This model is a fine-tuned version of Harveenchadha/vakyansh-wav2vec2-hindi-him-4200 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3192
• Wer: 0.3523

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: 0.0003
• 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: 10
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.3632	5.56	25	1.6683	0.5591
0.4727	11.11	50	1.4523	0.4156
0.3069	16.67	75	1.6365	0.3987
0.1225	22.22	100	1.4288	0.4367
0.0865	27.78	125	1.4239	0.4051
0.0496	33.33	150	1.3113	0.3819
0.0436	38.89	175	1.3556	0.3776
0.0337	44.44	200	1.2882	0.3586
0.0275	50.0	225	1.4075	0.3861
0.0204	55.56	250	1.3286	0.3734
0.016	61.11	275	1.3683	0.3861
0.0106	66.67	300	1.4088	0.3376
0.0096	72.22	325	1.3975	0.3565
0.0114	77.78	350	1.3268	0.3544
0.0078	83.33	375	1.3118	0.3439
0.0124	88.89	400	1.3192	0.3523

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1