โ† All examples

ModelEthio-ASR Amharic (600M)4Ethio-ASR Multi (300M)9Ethio-ASR Multi (600M)7Ethio-ASR Multi (94M)9Meta MMS-1B12OmniASR-CTC (1B)2OmniASR-CTC (300M)6OmniASR-CTC (3B)1OmniASR-LLM (1B)1OmniASR-LLM (300M)0OmniASR-LLM (3B)1OpenAI Whisper-small9
Audio qualityClean Studio8Ambient Room Noise3Harsh Environment4Distant / Low-Volume Mic7Wideband Phone (VoLTE)6Randomized Phone Channel4Narrowband 2G + Packet Loss8

What did the model get wrong?

OmniASR-CTC (300M) Wideband Phone (VoLTE)   Example 9 of 20  ยท  speaker: Female

๐ŸŽง Listen โ€” Wideband Phone (VoLTE)
Compressed, narrow-band audio like a regular phone call.
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
แ‰แŒฅแˆฉ แ‹œแˆฎแ‹˜แŒ แŠ แŠ แŠ•แ‹ต แˆแˆˆแ‰ต แŠ แˆแ‰ต แˆตแ‹ตแˆฐแ‰ฃแ‰ต แˆตแˆแŠ• แ‹˜แŒ แŠ แ‹œแˆฎ แАแ‹
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
6
Correct
4
Wrong
2
Missed
0
Extra
50%
Words right
Out of 12 spoken words, this model got 6 right (50%). Errors: 4 words wrong; 2 missed.

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