Armenian speech-to-text
PoorWe sent 30 Armenian test clips through this API. Whisper Large v3 made 50% word errors, which is not good enough to read as a transcript. This page shows what came back so you can judge for yourself.
Measured accuracy
| Model | Word error rate | Character error rate | Clips with no errors | Empty transcripts | Speed |
|---|---|---|---|---|---|
Whisper Large v3openai/whisper-large-v3 | 50% | 21% | 0 of 30 | 0 of 30 | 0.138× real time |
Poor. Good for topic and keywords. Not good enough to read as a transcript. The clips are clean, single-speaker read speech, so treat this as a best case for your own recordings. How we measured, and all 83 languages.
What a transcript looks like
The clip with the median error rate from the test, reference text against what the API returned.
| Spoken | Գիշերվա ընթացքում արվել են 150 և 200 պատճեներ, հիմա հայտնի են որպես «Դունլեպ լայնաշերտեր»: |
| Whisper Large v3 | Վիշերվա ընթացխում առվել են 150 և 200 պաճեներ, հիմա հայտնեն որպես դունլեպ լայնաշերտեր։ |
Language detection
All 10 clips we sent without a language field were identified as Armenian. Auto-detection is safe here; sending language=hy still skips the detection step.
What to do instead
We would rather you know this before you build than after. For Armenian, look at services that train specifically on the language, such as Google Cloud Speech-to-Text or Azure AI Speech, both of which publish their language lists, or a Whisper checkpoint fine-tuned on Armenian from Hugging Face run on your own hardware. If your audio mixes Armenian with a well-supported language, the well-supported parts will still transcribe. You can confirm all of this on your own audio with the free credit on a new account: transcription playground.
FAQ
How accurate is Whisper for Armenian?
On 30 clean read-speech clips it made 50% word errors, about one word in 2. Good for topic and keywords. Not good enough to read as a transcript. Real recordings with noise, accents or several speakers will score worse.
Does language auto-detection work for Armenian?
In our test, yes: all 10 clips sent without a language were identified as Armenian. Sending language=hy still saves the detection step on every request.
Can it translate Armenian speech to English?
The API returns Armenian text. For English, transcribe first and translate the text with a chat model on the same key; with accuracy this low the translation will inherit the errors.