metadata

language:
  - mn
license: apache-2.0
base_model: openai/whisper-large-v3
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_16_1
  - google/fleurs
metrics:
  - wer
model-index:
  - name: Whisper Large MN - Ankhbayasgalan Davaadorj
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 16.1 & FLEURS
          type: mozilla-foundation/common_voice_16_1
          config: mn
          split: None
          args: 'config: mn, split: test+validation'
        metrics:
          - name: Wer
            type: wer
            value: 33.65601452065343

Whisper Large MN - Ankhbayasgalan Davaadorj

This model is a fine-tuned version of openai/whisper-large-v3 on the Common Voice 16.1 & FLEURS dataset. It achieves the following results on the evaluation set:

Loss: 0.4942
Wer: 33.6560

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
0.0361	5.99	1000	0.3833	42.0109
0.0016	11.98	2000	0.4445	37.2092
0.0002	17.96	3000	0.4784	34.0410
0.0001	23.95	4000	0.4942	33.6560

Framework versions

Transformers 4.37.2
Pytorch 1.12.1+cu116
Datasets 2.17.0
Tokenizers 0.15.2