โ† All examples

ModelEthio-ASR Amharic (600M)4Ethio-ASR Multi (300M)8Ethio-ASR Multi (600M)4Ethio-ASR Multi (94M)8Meta MMS-1B12OmniASR-CTC (1B)1OmniASR-CTC (300M)3OmniASR-CTC (3B)1OmniASR-LLM (1B)1OmniASR-LLM (300M)1OmniASR-LLM (3B)1OpenAI Whisper-small19
Audio qualityClean Studio8Ambient Room Noise8Harsh Environment10Distant / Low-Volume Mic8Wideband Phone (VoLTE)9Randomized Phone Channel9Narrowband 2G + Packet Loss13

What did the model get wrong?

Ethio-ASR Multi (300M) Ambient Room Noise   Example 9 of 20  ยท  speaker: Female

๐ŸŽง Listen โ€” Ambient Room Noise
Realistic everyday background noise โ€” like a normal room or cafรฉ.
โœ… What was actually said
แ‰แŒฅแˆฉ แ‹œแˆฎ แ‹˜แŒ แŠ แŠ แŠ•แ‹ต แˆแˆˆแ‰ต แŠ แˆแˆตแ‰ต แˆตแ‹ตแˆตแ‰ต แˆฐแ‰ฃแ‰ต แˆตแˆแŠ•แ‰ต แ‹˜แŒ แŠ แ‹œแˆฎ แАแ‹
๐Ÿค– What this model heard
amh แ‰แŒฅแˆฉ แ‹œแˆฎแ‹˜แ‹˜แŒ แŠแŠ แŠ แŠ•แ‹ต แˆแˆˆแ‰ต แŠ แˆ แˆตแ‹ตแˆตแ‰ตแ‹ต แˆฐ แ‰ฃแ‰ต แˆตแˆแŠ•แ‹˜แ‹˜แŒ แŠ แ‹œแˆฎ แАแ‹
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
4
Correct
8
Wrong
0
Missed
0
Extra
33%
Words right
Out of 12 spoken words, this model got 4 right (33%). Errors: 8 words wrong.

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