| language: | |
| - luy | |
| license: apache-2.0 | |
| tags: | |
| - asr | |
| - automatic-speech-recognition | |
| - luhya | |
| - w2v-bert | |
| metrics: | |
| - wer | |
| - cer | |
| model-index: | |
| - name: luhya-asr-w2v-BERT | |
| results: | |
| - task: | |
| name: Automatic Speech Recognition | |
| type: automatic-speech-recognition | |
| metrics: | |
| - name: Test WER | |
| type: wer | |
| value: 43.4481 | |
| - name: Test CER | |
| type: cer | |
| value: 11.3584 | |
| # luhya-asr-w2v-BERT | |
| This model is a fine-tuned version of a w2v-BERT model for Automatic Speech Recognition in Luhya. | |
| ## Training results | |
| | Training Loss | Epoch | Step | Validation Loss | Wer | Cer | | |
| |:---:|:---:|:---:|:---:|:---:|:---:| | |
| | [More info] | 2.0 | [More info] | 0.4551 | 43.4481 | 11.3584 | | |
| ### Framework versions | |
| - Transformers 5.13.1 | |
| - Pytorch 2.11.0+cu128 | |
| - Datasets 4.0.0 | |
| - Tokenizers 0.22.2 | |