Dataset
Lwazi
ANV
← All examples
Model summary →
Model
facebook/omniASR-CTC-1B
Open
5
facebook/omniASR-CTC-300M
Open
6
facebook/omniASR-CTC-3B
Open
6
facebook/omniASR-LLM-1B
Open
4
facebook/omniASR-LLM-300M
Open
3
facebook/omniASR-LLM-3B
Open
2
facebook/omniASR-LLM-7B
Open
2
dsfsi/anv-whisper-large-v3-turbo-anv-zul
Open
0
dsfsi/anv-whisper-small-anv-zulu-first-batch
Open
4
openai/whisper-large-v3
Open
24
openai/whisper-small
Open
25
sitwala/whisper-large-anv-sot
Open
45
sitwala/whisper-large-v3-anv-sot
Open
52
sitwala/whisper-large-v3-turbo-anv-zul-150h
Open
3
sitwala/whisper-large-v3-turbo-anv-zul-250h
Open
1
sitwala/whisper-large-v3-turbo-anv-zul-50h
Open
4
What did the model get wrong?
openai/whisper-large-v3
Open
Example 48 of 100 · sample index 47
✅ What was actually said (isiZulu reference)
Ukuqasha
umhlaba
ngaphansi
kwezimo
ezivunayo
kumele
kubenze
bacabange
okukhulu,
hhayi
nje
ukukhulisa
amabhizinisi
abo
kodwa
ukwenza
ngcono
ingcebo
okwabelwana
ngayo
kanye
nokuchuma
emiphakathini
abalima
kuyona.
🤖 What this model heard
Ugo
kasha
umisaba
ngapanzi
gwezimo
ezifunali
kumele
kubenzi
batabangi
ukuhu.
Hai
nje
ugo
kulisa
ama
biznesi
abo
kotwa
uguenza
njono
indrebo
uguabelua
nangayo
kanyinu
kutuma
emipagatini
abalima
gionu.
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
4
Correct
21
Wrong
0
Missed
3
Extra
16%
Words right
0.960
WER
0.267
CER
24.57s
Duration
0.3s
Latency