LadislavVasina1/whisper-base-cs-cv11-timestretch20-gain10-pitch20-gaussian20-lowpass10-timemask10-freqmask10

whisper-base-cs-cv11-timestetch02-gain01-pitch02-gaussian02-lowpass01-timemask010-freqmask010

This model is a fine-tuned version of openai/whisper-base on the common_voice_11_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3005
• Wer: 30.8248

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 8000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.667	0.72	1000	0.4497	43.9367
0.506	1.44	2000	0.3698	37.1238
0.3637	2.17	3000	0.3355	34.1069
0.3686	2.89	4000	0.3198	32.7955
0.328	3.61	5000	0.3115	32.0938
0.2825	4.33	6000	0.3047	31.4086
0.2424	5.06	7000	0.3009	30.9555
0.2586	5.78	8000	0.3005	30.8248

Framework versions

• Transformers 4.38.1
• Pytorch 2.2.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2