metadata

language:
  - en
license: apache-2.0
base_model: openai/whisper-tiny
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: whispertiny-shreyas
    results: []

whispertiny-shreyas

This model is a fine-tuned version of openai/whisper-tiny on the AI4Bharat-svarah dataset. It achieves the following results on the evaluation set:

Loss: 0.5414
Wer: 22.8322

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: 2
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: 500
training_steps: 2000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2412	2.6702	1000	0.5319	22.8914
0.1071	5.3405	2000	0.5414	22.8322

Framework versions

Transformers 4.43.3
Pytorch 2.4.1
Datasets 2.14.7
Tokenizers 0.19.1