whisper-large-v3-Assamese-Version1

This model is a fine-tuned version of openai/whisper-large-v3 on the fleurs dataset. It achieves the following results on the evaluation set:

• Loss: 0.2353
• Wer: 62.8123

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-06
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• training_steps: 20000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3803	5.0505	2000	0.3681	78.7302
0.295	10.1010	4000	0.2985	71.4589
0.277	15.1515	6000	0.2724	68.1526
0.2493	20.2020	8000	0.2586	66.3248
0.2316	25.2525	10000	0.2492	64.9954
0.2236	30.3030	12000	0.2435	63.9927
0.2351	35.3535	14000	0.2401	63.2306
0.2089	40.4040	16000	0.2372	62.8295
0.2205	45.4545	18000	0.2358	62.5086
0.2253	50.5051	20000	0.2353	62.8123

Framework versions

• PEFT 0.12.1.dev0
• Transformers 4.45.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1