README.md

metadata

language:
  - sn
license: apache-2.0
library_name: peft
tags:
  - whisper-event and peft-lora
  - generated_from_trainer
base_model: openai/whisper-tiny
datasets:
  - Kittech/kittech_shona_dataset
metrics:
  - wer
model-index:
  - name: Whisper Small Sn - Bright Chirindo
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: Kingdom Truth International Ministries sermons
          type: Kittech/kittech_shona_dataset
          config: sn_zw
          split: test
          args: sn_zw
        metrics:
          - type: wer
            value: 208.88577256501785
            name: Wer

Whisper Small Sn - Bright Chirindo

This model is a fine-tuned version of kittech/whisper-tiny-sn on the Kingdom Truth International Ministries sermons dataset. It achieves the following results on the evaluation set:

• Loss: 2.9243
• Wer: 208.8858

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: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.5155	357.1429	2500	2.9250	212.4044
2.1292	714.2857	5000	2.9243	208.8858

Framework versions

• PEFT 0.10.1.dev0
• Transformers 4.40.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.2.dev0
• Tokenizers 0.19.1