deepdml/whisper-medium-ig-mix

Whisper Medium ig

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

• Loss: 1.5395
• Wer: 36.6214

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: 32
• 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: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1362	0.2	1000	1.2088	40.5087
0.0549	0.4	2000	1.3555	39.1381
0.0268	0.6	3000	1.4718	38.2932
0.0085	1.163	4000	1.5330	36.7742
0.0166	1.363	5000	1.5395	36.6214

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1

Citation

@misc{deepdml/whisper-medium-ig-mix,
      title={Fine-tuned Whisper medium ASR model for speech recognition in Igbo},
      author={Jimenez, David},
      howpublished={\url{https://huggingface.co/deepdml/whisper-medium-ig-mix}},
      year={2025}
    }