kannada_wav2vec2_final

This model is a fine-tuned version of Harveenchadha/vakyansh-wav2vec2-kannada-knm-560 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0031
• Wer: 0.1030

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.2326	8.33	25	0.2090	0.3620
0.2484	16.67	50	0.0718	0.1836
0.1391	25.0	75	0.0482	0.1308
0.0653	33.33	100	0.0091	0.1083
0.0616	41.67	125	0.0039	0.1044
0.0369	50.0	150	0.0031	0.1030

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 1.18.3
• Tokenizers 0.14.1