Hugging Face's logo

Mollel
/
ASR-Swahili-Small

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Swahili
whisper
Eval Results
Model card Files
xet
Metrics Community

Finetuned openai/whisper-small on 58000 Swahili training audio samples from mozilla-foundation/common_voice_17_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
Downloads last month
27
Safetensors
Model size
0.2B params
Tensor type
F32
·
Inference Providers NEW
Automatic Speech Recognition
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Mollel/ASR-Swahili-Small

Base model

openai/whisper-small
Finetuned
(3164)
this model

Dataset used to train Mollel/ASR-Swahili-Small

Evaluation results