README.md

metadata

language:
  - sw
license: apache-2.0
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_11_0
metrics:
  - wer
model-index:
  - name: Whisper Small Swahili
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 11.0
          type: mozilla-foundation/common_voice_11_0
          config: sw
          split: test
          args: 'config: sw, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 44.78584729981378

Whisper Small Swahili

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4630
• Wer: 44.7858

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
0.4	0.6	1000	0.5697	46.5687
0.2355	1.2	2000	0.4907	42.6466
0.2178	1.8	3000	0.4587	39.6855
0.1329	2.4	4000	0.4630	44.7858

Framework versions

• Transformers 4.28.0.dev0
• Pytorch 1.13.1+cu116
• Datasets 2.11.0
• Tokenizers 0.13.2