Dataset
Lwazi
ANV
← All examples
Model summary →
Model
facebook/omniASR-CTC-1B
Open
8
facebook/omniASR-CTC-300M
Open
7
facebook/omniASR-CTC-3B
Open
7
facebook/omniASR-LLM-1B
Open
7
facebook/omniASR-LLM-300M
Open
7
facebook/omniASR-LLM-3B
Open
6
facebook/omniASR-LLM-7B
Open
7
dsfsi/anv-whisper-large-v3-turbo-anv-zul
Open
4
dsfsi/anv-whisper-small-anv-zulu-first-batch
Open
5
openai/whisper-large-v3
Open
26
openai/whisper-small
Open
32
sitwala/whisper-large-anv-sot
Open
44
sitwala/whisper-large-v3-anv-sot
Open
44
sitwala/whisper-large-v3-turbo-anv-zul-150h
Open
3
sitwala/whisper-large-v3-turbo-anv-zul-250h
Open
4
sitwala/whisper-large-v3-turbo-anv-zul-50h
Open
4
What did the model get wrong?
openai/whisper-small
Open
Example 1 of 100 · sample index 0
✅ What was actually said (isiZulu reference)
Uma
idatha
yakho
yangonyaka
odlule
ikhombisa
uMhlaba
3
ngesivuno
esingaphansi
kwe-avareji
futhi,
ngakho
kufanele
sibheke
ngokuqaphela
esimo.
Mhlambe
uMhlaba
3
weswele
amanyuthriyenti
athile.
🤖 What this model heard
ḍmā
yītātāyāko
yānguñiāga
ozdole
yikonbisa
umla
bātre
yngesif
wūnno
esinga
pānti
wi'a
varej
fōti.
Yngāko
kufanel
e
si
bege
ngu
gukka
pēla
yi
sim.
Bslambe
umla
bātre
we'e
suele
āmanyu
triyendi
atī.
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
0
Correct
23
Wrong
0
Missed
9
Extra
0%
Words right
1.391
WER
0.447
CER
21.85s
Duration
0.3s
Latency