This model is a fine-tuned version of openai/whisper-small on the SPEAK-ASR/openslr-sinhala-asr-preprocessed-1 | SPEAK-ASR/openslr-sinhala-asr-preprocessed-2 | SPEAK-ASR/openslr-sinhala-asr-preprocessed-3 | SPEAK-ASR/youtube-sinhala-asr-preprocessed dataset. It achieves the following results on the evaluation set:

Loss: 0.1620
Wer: 21.2375

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-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
distributed_type: multi-GPU
num_devices: 4
total_train_batch_size: 128
total_eval_batch_size: 128
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: 1000
num_epochs: 15.0

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.3417	1.2755	1500	0.3163	35.3709
0.9804	2.5510	3000	0.2404	29.6283
0.8614	3.8265	4500	0.2112	26.8553
0.7851	5.1020	6000	0.1957	25.1669
0.7504	6.3776	7500	0.1856	24.1142
0.7110	7.6531	9000	0.1787	23.1621
0.6850	8.9286	10500	0.1731	22.7421
0.6610	10.2041	12000	0.1684	22.1571
0.6370	11.4796	13500	0.1652	21.7635
0.6355	12.7551	15000	0.1632	21.4154
0.6367	14.0306	16500	0.1620	21.2375

Framework versions

PEFT 0.18.1
Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.5.0
Tokenizers 0.22.2