โ† All examples

ModelEthio-ASR Amharic (600M)4Ethio-ASR Multi (300M)10Ethio-ASR Multi (600M)6Ethio-ASR Multi (94M)9Meta MMS-1B22OmniASR-CTC (1B)6OmniASR-CTC (300M)7OmniASR-CTC (3B)6OmniASR-LLM (1B)4OmniASR-LLM (300M)6OmniASR-LLM (3B)3OpenAI Whisper-small22
Audio qualityClean Studio7Ambient Room Noise12Harsh Environment10Distant / Low-Volume Mic7Wideband Phone (VoLTE)13Randomized Phone Channel17Narrowband 2G + Packet Loss20

What did the model get wrong?

Ethio-ASR Multi (300M) Harsh Environment   Example 2 of 20  ยท  speaker: Male

๐ŸŽง Listen โ€” Harsh Environment
Very heavy background noise โ€” much harder than typical audio.
โœ… What was actually said
แ‰ฃแˆ…แˆญ แ‹ณแˆญ แАแ‹ แ‹ซแˆˆแˆแ‰ต แŠขแŠ•แ‰ฐแˆญแŠ”แ‰ต แ‰ฐแ‰‹แˆญแŒฆแ‰ฅแŠ›แˆ แ‰ฅแˆแˆฝแ‰ต แˆ›แˆตแˆ˜แ‹แŒˆแ‰ฅ แˆแˆแŒŒ แАแ‰ แˆญ แˆตแˆแŠฌ แ‹œแˆฎ แ‹˜แŒ แŠ แŠ แŠ•แ‹ต แŠ แŠ•แ‹ต แˆตแ‹ตแˆตแ‰ต แˆฐแ‰ฃแ‰ต แ‹œแˆฎ แˆแˆˆแ‰ต แ‹˜แŒ แŠ แŠ แˆแˆตแ‰ต แАแ‹
๐Ÿค– What this model heard
amh แ‰ฃแˆ…แˆญ แ‹ณแˆญ แАแ‹ แ‹ซแˆˆแ‹แ‰ต แ‰ดแˆŒแ‰ฅแˆญ แ‰ฐแ‰‹แˆญแŒฆแ‰ฅแŠ›แˆ แ‰ฅแˆแˆฝแ‰ต แˆ›แˆตแˆ˜แˆตแŒˆแ‰ฅ แŠ แˆแ‰… แ‹จแАแ‰ แˆญ แˆตแˆแŠฌ แ‹œแ‹œแˆฎ แ‹˜แ‹˜แŒ แŠแŠ แŠ แŠ•แ‹ต แˆตแ‹ตแˆตแ‰ต แˆฐแ‰ฃแ‰ต แ‹œแˆฎ แˆแˆˆแ‰ต แ‹˜แ‹˜แŒ แŠแŠ แŠ แˆแˆตแ‰ต แАแ‹
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
13
Correct
8
Wrong
1
Missed
1
Extra
59%
Words right
Out of 22 spoken words, this model got 13 right (59%). Errors: 8 words wrong; 1 missed; 1 extra.

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