futureProofGlitch/whisper-small

Whisper Small - FutureProofGlitch

This model is a fine-tuned version of openai/whisper-small on the AMI Meeting Corpus dataset. It achieves the following results on the evaluation set:

• Loss: 0.4325
• Wer Ortho: 19.5838
• Wer: 19.3832

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: 8
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 50
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.2735	0.61	500	0.3324	21.5310	21.2081
0.1235	1.22	1000	0.3473	19.6819	19.4991
0.1317	1.83	1500	0.3342	19.0920	18.7929
0.0647	2.44	2000	0.3671	22.8615	22.6949
0.0294	3.05	2500	0.3842	18.5566	18.4101
0.0534	3.66	3000	0.4044	20.8094	20.5998
0.0366	4.27	3500	0.4277	20.2686	20.1372
0.0328	4.88	4000	0.4325	19.5838	19.3832

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2