goaicorp/mos-whisper-large-v3-turbo

whisper Large V3 Turbo for Mooré - GO AI CORP

This model is a fine-tuned version of openai/whisper-large-v3-turbo on the moore-tts-full-dataset dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.4706
• eval_wer: 36.7122
• eval_runtime: 698.1717
• eval_samples_per_second: 3.482
• eval_steps_per_second: 0.218
• epoch: 5.1170
• step: 7000

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: 16
• seed: 42
• optimizer: Use 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: 500
• training_steps: 10000
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.46.2
• Pytorch 2.2.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3