โ† All examples

ModelEthio-ASR Amharic (600M)8Ethio-ASR Multi (300M)18Ethio-ASR Multi (600M)9Ethio-ASR Multi (94M)9Meta MMS-1B9OmniASR-CTC (1B)9OmniASR-CTC (300M)8OmniASR-CTC (3B)4OmniASR-LLM (1B)8OmniASR-LLM (300M)8OmniASR-LLM (3B)9OpenAI Whisper-small9
Audio qualityClean Studio4Ambient Room Noise2Harsh Environment4Distant / Low-Volume Mic4Wideband Phone (VoLTE)3Randomized Phone Channel3Narrowband 2G + Packet Loss9

What did the model get wrong?

OmniASR-CTC (1B) Narrowband 2G + Packet Loss   Example 7 of 20  ยท  speaker: Female

๐ŸŽง Listen โ€” Narrowband 2G + Packet Loss
Heavy background noise plus phone-call compression โ€” worst-case stress test.
โœ… What was actually said
แ‹จแŠขแŠ•แ‰ฐแˆญแŠ”แ‰ต แŒฅแ‰…แˆ แ‹ˆแ‹ญแˆ แ“แŠฌแŒ… แˆ˜แŒแ‹›แ‰ต แŠฅแ‰ฝแˆ‹แˆˆแˆ แ‹‹แŒ‹แ‹ แˆตแŠ•แ‰ต แŠแ‹
๐Ÿค– What this model heard
inte nite kabie fakic nag sakilano yagaitne
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
0
Correct
7
Wrong
2
Missed
0
Extra
0%
Words right
Out of 9 spoken words, this model got 0 right (0%). Errors: 7 words wrong; 2 missed.

Try switching Audio quality โ€” see the same model degrade as the audio gets noisier.