โ† All examples

ModelEthio-ASR Amharic (600M)3Ethio-ASR Multi (300M)15Ethio-ASR Multi (600M)3Ethio-ASR Multi (94M)5Meta MMS-1B7OmniASR-CTC (1B)7OmniASR-CTC (300M)7OmniASR-CTC (3B)7OmniASR-LLM (1B)2OmniASR-LLM (300M)6OmniASR-LLM (3B)5OpenAI Whisper-small21
Audio qualityClean Studio7Ambient Room Noise17Harsh Environment19Distant / Low-Volume Mic7Wideband Phone (VoLTE)7Randomized Phone Channel7Narrowband 2G + Packet Loss21

What did the model get wrong?

OpenAI Whisper-small Narrowband 2G + Packet Loss   Example 3 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‰ แˆตแˆแŠญ แ‹ฐแ‹แ‹จ แˆ›แŒแŠ˜แ‰ต แ‹จแˆแ‰ฝแˆแ‰ แ‰ต แ‹จแŒฅแˆช แˆ›แŠฅแŠจแˆ แŠ แˆˆ
๐Ÿค– What this model heard
bf el kita wiz mimany tum mikh hul kait s y e t ar im a a k an an le
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
0
Correct
7
Wrong
0
Missed
14
Extra
0%
Words right
Out of 7 spoken words, this model got 0 right (0%). Errors: 7 words wrong; 14 extra.

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