Dataset
Lwazi
ANV
← All examples
Model summary →
Model
facebook/omniASR-CTC-1B
Open
7
facebook/omniASR-CTC-300M
Open
8
facebook/omniASR-CTC-3B
Open
4
facebook/omniASR-LLM-1B
Open
9
facebook/omniASR-LLM-300M
Open
6
facebook/omniASR-LLM-3B
Open
6
facebook/omniASR-LLM-7B
Open
5
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
10
openai/whisper-small
Open
20
sitwala/whisper-large-anv-sot
Open
23
sitwala/whisper-large-v3-anv-sot
Open
24
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?
sitwala/whisper-large-anv-sot
Open
Example 16 of 100 · sample index 15
✅ What was actually said (isiZulu reference)
UMose,
Greek,
Moishe,
Yiddish,
Moshe,
Hebrew,
noma
iMovses,
Armenian,
yigama
elinikezwe
owesilisa,
ngemuva
kwesibalo
seBhayibheli
uMose.
🤖 What this model heard
o
mose
creek
mmoeshe
yiddish
mmoshe
hebrew
noma
in
moses
armenian
ye
kama
e
le
nekezwe
o
eselesa
nge
mmova
ho
e
sepalo
sepae
pele
o
mose
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
4
Correct
12
Wrong
0
Missed
11
Extra
25%
Words right
1.625
WER
0.346
CER
20.54s
Duration
0.3s
Latency