DatasetLwaziANV

← All examples Model summary →

Modelfacebook/omniASR-CTC-1BOpen2facebook/omniASR-CTC-300MOpen3facebook/omniASR-CTC-3BOpen1facebook/omniASR-LLM-1BOpen2facebook/omniASR-LLM-300MOpen2facebook/omniASR-LLM-3BOpen2facebook/omniASR-LLM-7BOpen2dsfsi/anv-whisper-large-v3-turbo-anv-zulOpen4dsfsi/anv-whisper-small-anv-zulu-first-batchOpen4openai/whisper-large-v3Open31openai/whisper-smallOpen19sitwala/whisper-large-anv-sotOpen42sitwala/whisper-large-v3-anv-sotOpen41sitwala/whisper-large-v3-turbo-anv-zul-150hOpen4sitwala/whisper-large-v3-turbo-anv-zul-250hOpen4sitwala/whisper-large-v3-turbo-anv-zul-50hOpen5

What did the model get wrong?

facebook/omniASR-LLM-3BOpen   Example 3 of 100  ·  sample index 2

✅ What was actually said (isiZulu reference)
Kubalulekileukwenzaikhalibhreshinieqondileyokwendlalaumanyolongaphambikokubauqaleukuthela.Kubalulekileukubautheleumanyoloonqunyiweukuzeutholeumphumelaophezulu.
🤖 What this model heard
kubalulekileukwenzaicolibrationeqondileyokwendlalaumanyolongaphambikokubauqaleukuthelakubalulekileukubautheleumanyoloonqunyiweukuzeutholeumphumelaophezulu
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
18
Correct
1
Wrong
0
Missed
1
Extra
95%
Words right
0.211
WER
0.062
CER
19.01s
Duration
1.5s
Latency