Dataset
Lwazi
ANV
← All examples
Model summary →
Model
facebook/omniASR-CTC-1B
Open
4
facebook/omniASR-CTC-300M
Open
4
facebook/omniASR-CTC-3B
Open
2
facebook/omniASR-LLM-1B
Open
4
facebook/omniASR-LLM-300M
Open
4
facebook/omniASR-LLM-3B
Open
3
facebook/omniASR-LLM-7B
Open
3
dsfsi/anv-whisper-large-v3-turbo-anv-zul
Open
5
dsfsi/anv-whisper-small-anv-zulu-first-batch
Open
6
openai/whisper-large-v3
Open
30
openai/whisper-small
Open
148
sitwala/whisper-large-anv-sot
Open
45
sitwala/whisper-large-v3-anv-sot
Open
45
sitwala/whisper-large-v3-turbo-anv-zul-150h
Open
5
sitwala/whisper-large-v3-turbo-anv-zul-250h
Open
3
sitwala/whisper-large-v3-turbo-anv-zul-50h
Open
5
What did the model get wrong?
sitwala/whisper-large-anv-sot
Open
Example 34 of 100 · sample index 33
✅ What was actually said (isiZulu reference)
Ngaphezu
kwalokho,
izinyoni
nezinye
izilwane
zizalela
amaqanda,
okuvame
ukudliwa,
futhi
izinyosi
zikhiqiza
uju,
umpe
olincishisiwe
oluqhamuka
ezimbalini,
okuyi-sweetener
ethandwayo
emasikweni
amaningi.
🤖 What this model heard
ngaphezo
wa
logo
e
zenyone
ne
zene
e
zelwane
se
zalela
a
makae
o
bovame
o
lelewa
fothi
e
zenyo
se
zikikreza
o
botso
o
mpe
o
leng
nshe
sewe
o
loka
moga
e
zembaleng
o
bo
e
soetna
e
tshandwaeo
e
masegwene
a
maneng
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
0
Correct
21
Wrong
0
Missed
24
Extra
0%
Words right
2.143
WER
0.473
CER
25.38s
Duration
0.3s
Latency