Zum Inhalt springen
ModelAufgelistetExtern
Herunterladen

Whisper Small SIT 16 07 26 0457

This model is a fine-tuned version of Pengwin30/whisper-small-fine-tuned on an unknown dataset. It achieves the following results on the evaluation set: Loss: 0.0110 Wer: 0.4639

ArtText modelBasismodellPengwin30/whisper-small-fine-tunedVersionFP32BBewertung

Modellquelle

Art
Text model
Dateigröße
922 MB
Basismodell
Pengwin30/whisper-small-fine-tuned
Version
vFP32
Aktualisiert
31. Okt. 2026
Downloads
7
Quelle
Hugging Face

Quellenbeschreibung

whisper-small-SIT_16_07_26_0457

This model is a fine-tuned version of Pengwin30/whisper-small-fine-tuned on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0110
  • Wer: 0.4639

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: 4
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 16
  • optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepValidation LossWer
0.75781.01810.02620.8615
0.01562.03620.01320.3976
0.00503.05430.00700.3976
0.00054.07240.01100.4639
0.00015.09050.01100.4639

Framework versions

  • Transformers 5.12.1
  • Pytorch 2.11.0+cu128
  • Datasets 2.19.0
  • Tokenizers 0.22.2

Quellen

1 Quelle
Hugging FacePrimärVersion 0a3dde9976162c6563c1ef68fb7c597f0f32c33c
huggingface.co/horhh/whisper-small-SIT_16_07_26_0457/tree/0a3dde9976162c6563c1ef68fb7c597f0f32c33c

Verifiziert 31. Okt.

Download files

Modellartefakte

1 Artefakt

model.safetensors

safetensors · FP32 (unverified) · 922 MB · SHA-256 340b8f0f4be1…47af · Hugging Face

Download file

Quellenauszüge

3 Auszüge

--- library_name: transformers license: mit base_model: Pengwin30/whisper-small-fine-tuned tags: - generated_from_trainer metrics: - wer model-index: - name: whisper-small-SIT_16_07_26_0457 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # whisper-small-SIT_16_07_26_0457 This model is a fine-tuned version of [Pengwin30/whisper-small-fine-tuned](https://huggingface.co/Pengwin30/whisper-small-fine-tuned) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.0110 - Wer: 0.4639 ## 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: 4 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 16 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 100 - num_epochs: 5 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:------:| | 0.7578 | 1.0 | 181 | 0.0262 | 0.8615 | | 0.0156 | 2.0 | 362 | 0.0132 | 0.3976 | | 0.0050 | 3.0 | 543 | 0.0070 | 0.3976 | | 0.0005 | 4.0 | 724 | 0.0110 | 0.4639 | | 0.0001 | 5.0 | 905 | 0.0110 | 0.4639 | ### Framework versions - Transformers 5.12.1 - Pytorch 2.11.0+cu128 - Datasets 2.19.0 - Tokenizers 0.22.2

Source context: 7 downloads · 0 likes · Pipeline automatic-speech-recognition · Library transformers · Repo horhh/whisper-small-SIT_16_07_26_0457

Source context: 16 downloads · 0 likes · Pipeline automatic-speech-recognition · Library transformers · Repo horhh/whisper-small-SIT_16_07_26_0457