HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

whisper-large-v3-swahili-afrivoice-cv

ElizabethMwangi/whisper-large-v3-swahili-afrivoice-cv

kind
original
quant
bf16 (16-bit)
family
misc
modality
automatic-speech-recognition
root model
openai/whisper-large-v3
provider
ElizabethMwangi
license
apache-2.0
language
sw
context
—
library
—
vRAM (GB)
—
size on disk
—
downloads
1,668
likes
0
created
—
last modified
2026-06-15T09:16:11+00:00 (2026)
huggingface
https://huggingface.co/ElizabethMwangi/whisper-large-v3-swahili-afrivoice-cv
tags
safetensorswhisperautomatic-speech-recognitionswahilikenyalow-resourceswdataset:ElizabethMwangi/swahili_afrivoice_all_domains_v0.1dataset:cdli/common_voice_swahili_smallbase_model:openai/whisper-large-v3base_model:finetune:openai/whisper-large-v3license:apache-2.0region:us

Benchmark scores (3)

benchmarkcategoryscorecontextconfevidence
WERspeech-asr23.93mediumWER 0.2393
WER Common Voicespeech-asr20.1CommonVoice Swahili testmedium| CommonVoice Swahili test | 0.201 | 0.049 |
CERspeech-asr8.9low| AfriVoice dev | 0.239 | 0.089 |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.