This model is a fine-tuned version of openai/whisper-small on the ne-en-codeswitching-asr-technical-interview dataset. It achieves the following results on the evaluation set:

Loss: 0.4139
Wer: 32.1763

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: 1e-05
train_batch_size: 32
eval_batch_size: 16
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 50
training_steps: 500
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.4745	2.1739	50	0.7265	49.7521
0.5425	4.3478	100	0.4280	35.2066
0.1929	6.5217	150	0.3774	33.9945
0.0776	8.6957	200	0.3845	33.0028
0.0324	10.8696	250	0.3955	36.7493
0.0125	13.0435	300	0.3997	33.6088
0.0063	15.2174	350	0.4029	32.5620
0.0045	17.3913	400	0.4103	32.2865
0.0031	19.5652	450	0.4123	32.2314
0.0032	21.7391	500	0.4139	32.1763

Framework versions

Transformers 5.2.0
Pytorch 2.9.0+cu126
Datasets 4.0.0
Tokenizers 0.22.2

e-horizon/whisper-small-ne-en

Whisper Small Code Switching Ne - Saurav Karki

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

e-horizon/
whisper-small-ne-en