Dataset
Lwazi
ANV
← All examples
Model summary →
Model
facebook/omniASR-CTC-1B
Open
10
facebook/omniASR-CTC-300M
Open
11
facebook/omniASR-CTC-3B
Open
10
facebook/omniASR-CTC-7B
Open
11
facebook/omniASR-LLM-1B
Open
10
facebook/omniASR-LLM-300M
Open
9
facebook/omniASR-LLM-3B
Open
9
facebook/omniASR-LLM-7B
Open
10
aadel4/omniASR-CTC-1B-v2
Open
10
aadel4/omniASR-CTC-300M-v2
Open
11
badrex/w2v-bert-2.0-zulu-asr
Open
10
dsfsi/anv-whisper-large-v3-turbo-anv-zul
Open
11
dsfsi/anv-whisper-small-anv-zulu-first-batch
Open
11
facebook/mms-1b-all
Open
12
openai/whisper-large-v3
Open
17
openai/whisper-small
Open
16
sitwala/whisper-large-anv-sot
Open
21
sitwala/whisper-large-v3-anv-sot
Open
203
sitwala/whisper-large-v3-turbo-anv-zul-150h
Open
11
sitwala/whisper-large-v3-turbo-anv-zul-250h
Open
10
sitwala/whisper-large-v3-turbo-anv-zul-50h
Open
10
What did the model get wrong?
aadel4/omniASR-CTC-300M-v2
Open
Example 59 of 100 · sample index 58
✅ What was actually said (isiZulu reference)
kwathi
u-
Israel
esenama-
esenamandla,
[n]
wawasebenzisa,
[n]
[s]
amaKhanani
kepha,
[n]
angiboni-ke
ma-
akawaxoshanga
nokuwaxosha.
🤖 What this model heard
kwathi
u-israyeli
esenamandla
wabasebenzisa
amakhanani
kepha
angibonikexoshwanga
nokuwaxosha
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
5
Correct
3
Wrong
8
Missed
0
Extra
31%
Words right
0.833
WER
0.246
CER
15.73s
Duration
0.0s
Latency