โ† All examples

ModelEthio-ASR Amharic (600M)3Ethio-ASR Multi (300M)9Ethio-ASR Multi (600M)8Ethio-ASR Multi (94M)8Meta MMS-1B12OmniASR-CTC (1B)2OmniASR-CTC (300M)4OmniASR-CTC (3B)1OmniASR-LLM (1B)0OmniASR-LLM (300M)0OmniASR-LLM (3B)1OpenAI Whisper-small19
Audio qualityClean Studio12Ambient Room Noise19Harsh Environment217Distant / Low-Volume Mic16Wideband Phone (VoLTE)9Randomized Phone Channel19Narrowband 2G + Packet Loss12

What did the model get wrong?

OpenAI Whisper-small Randomized Phone Channel   Example 9 of 20  ยท  speaker: Female

๐ŸŽง Listen โ€” Randomized Phone Channel
Phone-call audio with random extra distortions layered on top.
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 แŠแ‹
๐Ÿค– What this model heard
แ‹œแˆฎ because zero is zero and let s add the red sauce and let s add the red sauce
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
0
Correct
12
Wrong
0
Missed
7
Extra
0%
Words right
Out of 12 spoken words, this model got 0 right (0%). Errors: 12 words wrong; 7 extra.

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