elliottower1/whisper-small-id

Whisper Small ID - Common Voice 17

This model is a fine-tuned version of openai/whisper-small on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.2892
• eval_wer: 17.9219
• eval_runtime: 1041.4909
• eval_samples_per_second: 3.496
• eval_steps_per_second: 0.438
• epoch: 3.8462
• step: 2000

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.44.2
• Pytorch 2.3.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1

Training Environment

This model was trained on a single A100 GPU machine in Google Cloud. Below are the machine specifications:

Machine Type	GPU Count	GPU Memory (GB HBM2)	vCPU Count	VM Memory (GB)	Local SSD Supported	Max Network Bandwidth (Gbps)
a2-highgpu-1g	1	40	12	85	Yes	24

You can find more details about the machine type here.

Training Results

Training Loss	Step	Validation Loss	Wer
0.2128	1000	0.251406	17.495011
0.0270	2000	0.289191	17.921945