CKSINGH/whisper-small-hi-firefox

CKSINGH whisper-small-hi-fiefox

This model is a fine-tuned version of CKSINGH/whisper-small-hi-iiib on the Common Voice 13.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2473
• Wer: 11.4253
• Cer: 7.5700

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: 1.75e-05
• train_batch_size: 16
• 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: 500
• training_steps: 3000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.0382	3.57	1000	0.2021	12.8043	8.1173
0.0027	7.14	2000	0.2321	11.6614	7.6820
0.0001	10.71	3000	0.2473	11.4253	7.5700

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0