This method is a finetuned version of openai/whisper-large-v2 using PEFT and Lora. The model achieved a WER score of 21% and a CER of 0.088 on the test split of djelia/bambara-asr
- Downloads last month
- 2
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for djelia/bm-whisper-large-v2-lora-merged
Base model
openai/whisper-large-v2