Mollel/ASR-Swahili-Small

Finetuned openai/whisper-small on 58000 Swahili training audio samples from mozilla-foundation/commonvoice17_0.

This model was created from the Mozilla.ai Blueprint: speech-to-text-finetune.

Evaluation results on 12253 audio samples of Swahili:

Baseline model (before finetuning) on Swahili

• —Word Error Rate: 133.795
• —Loss: 2.459

Finetuned model (after finetuning) on Swahili

• —Word Error Rate: 43.876
• —Loss: 0.653