Languages: Swahili (sw) โ€” Non-standard / accented varieties from Kenyan speakers

This model was fine-tuned on non-standard and accented Swahili speech using Common Voice and additional curated speech data from Kenyan speakers. The goal was to improve ASR performance on regional Swahili accents and non-standard pronunciations.

Fine-tuning data: Subset of Mozilla Common Voice + curated non-standard Swahili speech

๐Ÿ“Š Evaluation

Metric Pre-train Test
WER 0.3737 0.3123
CER 0.1566 0.1317
Downloads last month
130
Safetensors
Model size
0.2B params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for Veronica1NW/sw_whisper_nonstandard_small

Finetuned
(3003)
this model