README.md

---
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-large-v2-sw
    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: 30.7
---
## Model
* Name: Whisper Large-v2 Swahili
* Description: Whisper weights for speech-to-text task, fine-tuned and evaluated on normalized data.
* Dataset:
  - Train and validation splits for Swahili subsets of [Common Voice 11.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_11_0).
  - Train, validation and test splits for Swahili subsets of [Google Fleurs](https://huggingface.co/datasets/google/fleurs/). 
* Performance: **30.7 WER**

## Weights
* Date of release: 12.09.2022
* License: MIT

## Usage
To use these weights in HuggingFace's `transformers` library, you can do the following:
```python
from transformers import WhisperForConditionalGeneration

model = WhisperForConditionalGeneration.from_pretrained("hedronstone/whisper-large-v2-sw")
```