./whisper-large-cit-synth-do0.15-wd0-lr1e-05

This model is a fine-tuned version of openai/whisper-large-v3 on the SF 200 synth 2000 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3455
• Wer: 15.0359

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: 4
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• training_steps: 300
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.5226	0.4040	50	0.3027	17.5235
0.2202	0.8081	100	0.2734	16.6022
0.1105	1.2121	150	0.2876	16.1047
0.0613	1.6162	200	0.2642	14.3542
0.0504	2.0202	250	0.3025	14.3910
0.0158	2.4242	300	0.3455	15.0359

Framework versions

• Transformers 4.41.2
• Pytorch 1.13.1+cu117
• Datasets 2.19.2
• Tokenizers 0.19.1