iulik-pisik's picture
End of training
97847b2 verified
metadata
language:
  - ro
license: apache-2.0
base_model: openai/whisper-tiny
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - iulik-pisik/horoscop_neti
metrics:
  - wer
model-index:
  - name: Whisper Tiny Romanian - Horoscop Neti
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Horoscop Neti
          type: iulik-pisik/horoscop_neti
          config: default
          split: None
          args: 'config: ro, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 21.854465806765635

Whisper Tiny Romanian - Horoscop Neti

This model is a fine-tuned version of openai/whisper-tiny on the Horoscop Neti dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4305
  • Wer: 21.8545

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: 4000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.0401 9.71 1000 0.3338 22.5116
0.0027 19.42 2000 0.3979 22.5359
0.0013 29.13 3000 0.4230 22.0248
0.001 38.83 4000 0.4305 21.8545

Framework versions

  • Transformers 4.38.1
  • Pytorch 2.1.0+cu121
  • Datasets 2.17.1
  • Tokenizers 0.15.2