UDA-LIDI/openai-whisper-large-v3-fullFT-es_ecu911_V2martin_win30s_samples

Whisper openai-whisper-large-v3

This model is a fine-tuned version of openai/whisper-large-v3 on the llamadas ecu911 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2501
• Wer: 19.6292

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: 2e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.5057	1.0	422	0.3380	35.0250
0.2061	2.0	844	0.2458	32.0536
0.1136	3.0	1266	0.2270	26.1110
0.0685	4.0	1688	0.2281	17.5388
0.0444	5.0	2110	0.2248	18.6169
0.0301	6.0	2532	0.2470	18.6037
0.0234	7.0	2954	0.2420	18.1699
0.019	8.0	3376	0.2368	21.5751
0.0163	9.0	3798	0.2346	15.9216
0.0142	10.0	4220	0.2501	19.6292

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1