Hugging Face's logo

Eun0503
/
safecall-model

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Korean
whisper
hf-asr-leaderboard
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
safecall-model / README.md
Eun0503's picture
Eun0503
End of training
a8222b0 verified
preview code
|
raw
history blame
No virus
1.56 kB
metadata
language:
  - ko
license: apache-2.0
base_model: openai/whisper-base
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - Eun0503/safecall-preprocessed
model-index:
  - name: 음성인식 for safecall
    results: []

음성인식 for safecall

This model is a fine-tuned version of openai/whisper-base on the 저음질 전화망 음성인식 데이터 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4638
  • Cer: 54.7929

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 1000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Cer
0.0296 10.0 1000 0.4638 54.7929

Framework versions

  • Transformers 4.41.0.dev0
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1