โ† All examples

ModelEthio-ASR Amharic (600M)7Ethio-ASR Multi (300M)2Ethio-ASR Multi (600M)5Ethio-ASR Multi (94M)4Meta MMS-1B12OmniASR-CTC (1B)0OmniASR-CTC (300M)2OmniASR-CTC (3B)0OmniASR-LLM (1B)1OmniASR-LLM (300M)0OmniASR-LLM (3B)0OpenAI Whisper-small12
Audio qualityClean Studio0Ambient Room Noise0Harsh Environment0Distant / Low-Volume Mic0Wideband Phone (VoLTE)0Randomized Phone Channel1Narrowband 2G + Packet Loss8

What did the model get wrong?

OmniASR-LLM (3B) Harsh Environment   Example 18 of 20  ยท  speaker: Female

๐ŸŽง Listen โ€” Harsh Environment
Very heavy background noise โ€” much harder than typical audio.
โœ… What was actually said
แ‰แŒฅแˆฌ แ‹œแˆฎ แ‹˜แŒ แŠ แŠ แˆแˆตแ‰ต แŠ แŠ•แ‹ต แŠ แŠ•แ‹ต แˆตแˆแŠ•แ‰ต แŠ แˆแˆตแ‰ต แŠ แŠ•แ‹ต แŠ แˆซแ‰ต แŠ แŠ•แ‹ต แŠแ‹
๐Ÿค– What this model heard
แ‰แŒฅแˆฌ แ‹œแˆฎ แ‹˜แŒ แŠ แŠ แˆแˆตแ‰ต แŠ แŠ•แ‹ต แŠ แŠ•แ‹ต แˆตแˆแŠ•แ‰ต แŠ แˆแˆตแ‰ต แŠ แŠ•แ‹ต แŠ แˆซแ‰ต แŠ แŠ•แ‹ต แŠแ‹
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
12
Correct
0
Wrong
0
Missed
0
Extra
100%
Words right
Out of 12 spoken words, this model got 12 right (100%). Errors: everything correct.

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