cmuru39
/

whisper_large_swahili

Automatic Speech Recognition

Model card Files Files and versions

Trained on CDLI's non-standard speech dataset (Kenyan Swahili): cdli/kenyan_swahili_nonstandard_speech_v0.9

Performance on test set before and after:

before: WER on test: 0.41
after: WER on test: 0.24

More details: https://www.cdl-inclusion.com/ASRKenya

Downloads last month: 32

Safetensors

Model size

2B params

Tensor type

F32

·

Model tree for cmuru39/whisper_large_swahili

Base model

openai/whisper-small

Finetuned

(3515)

this model