Whisper Small Lithuanian and Serbian sequentially trained

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

Lithuanian

• Wer: >100

Serbian

• Wer: 35.6131

Training procedure

It was first trained 2000 steps on Lithuanian and then 2000 steps on Serbian, continuing from the last checkpoint for Lithuanian.

Training hyperparameters per fine-tune

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 64
• 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: 2000
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0+cu117
• Datasets 2.7.1.dev0
• Tokenizers 0.13.2