cdli-whisper-ml-lug-swa-sunbird-full-a40-lr1e4 – AI Model by KasuleTrevor | AlphaNeural AI

You can deploy this model and start earning money today!

KasuleTrevor/
cdli-whisper-ml-lug-swa-sunbird-full-a40-lr1e4

like

0

automatic-speech-recognition

generated_from_trainer

Sunbird/asr-whisper-large-v3-salt

endpoints_compatible

Views

No views yet

cdli-whisper-ml-lug-swa-sunbird-full-a40-lr1e4

This model is a fine-tuned version of Sunbird/asr-whisper-large-v3-salt on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.8696
Wer: 0.4548
Cer: 0.2149

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: 0.0001
train_batch_size: 4
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
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: 150
training_steps: 2500

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.8765	0.8068	500	1.0419	0.5116	0.2574
0.5085	1.6132	1000	0.8285	0.4775	0.2382
0.2525	2.4195	1500	0.8308	0.4671	0.2263
0.1164	3.2259	2000	0.8705	0.4478	0.2138
0.092	4.0323	2500	0.8696	0.4548	0.2149

Framework versions

Transformers 4.52.0
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.21.4