Dataset
Lwazi
ANV
← All examples
Model summary →
Model
facebook/omniASR-CTC-1B
Open
11
facebook/omniASR-CTC-300M
Open
13
facebook/omniASR-CTC-3B
Open
11
facebook/omniASR-CTC-7B
Open
7
facebook/omniASR-LLM-1B
Open
7
facebook/omniASR-LLM-300M
Open
10
facebook/omniASR-LLM-3B
Open
10
facebook/omniASR-LLM-7B
Open
7
aadel4/omniASR-CTC-1B-v2
Open
13
aadel4/omniASR-CTC-300M-v2
Open
12
badrex/w2v-bert-2.0-zulu-asr
Open
8
dsfsi/anv-whisper-large-v3-turbo-anv-zul
Open
7
dsfsi/anv-whisper-small-anv-zulu-first-batch
Open
10
facebook/mms-1b-all
Open
14
openai/whisper-large-v3
Open
15
openai/whisper-small
Open
23
sitwala/whisper-large-anv-sot
Open
19
sitwala/whisper-large-v3-anv-sot
Open
18
sitwala/whisper-large-v3-turbo-anv-zul-150h
Open
8
sitwala/whisper-large-v3-turbo-anv-zul-250h
Open
8
sitwala/whisper-large-v3-turbo-anv-zul-50h
Open
9
What did the model get wrong?
facebook/omniASR-CTC-3B
Open
Example 44 of 100 · sample index 43
✅ What was actually said (isiZulu reference)
[s]
iqanda,
i-
isikhombisa,
isishiyagalombili,
isishiyagalombili,
iqanda,
kuthathu,
kune,
iqanda,
kuhlanu,
bese
kuba
elinye
iqanda.
🤖 What this model heard
indilinga
i
isikhombisa
isishiyaga
lombili
isishiyagalombili
indilinga
thathu
kune
indilinga
kuhlanu
besekuba
yinye
indilinga
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
5
Correct
8
Wrong
2
Missed
1
Extra
33%
Words right
1.000
WER
0.323
CER
18.22s
Duration
0.1s
Latency