Hypa_Whisper STT
Collection
A family of fine-tuned Whisper models by Hypa AI, built for robust, low-resource, multilingual automatic speech recognition with support for diverse s • 2 items • Updated
• 1
wspr_wazobia_run2_04282025
This model is a fine-tuned version of openai/whisper-small on an unknown dataset.
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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- training_steps: 7000
- mixed_precision_training: Native AMP
Training results
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.1
- Tokenizers 0.21.1
- Downloads last month
- 350
Model tree for hypaai/hypaai-whisper-small-v2-04282025
Base model
openai/whisper-small