Nepali speech-to-text
Not usableWe sent 30 Nepali test clips through this API. Whisper Large v3 made 92% 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 | 92% | 33% | 0 of 30 | 2 of 30 | 0.114× real time |
Not usable. Most words come out wrong. Do not build on it for this language. 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 | उहाँले “उनीहरूलाई आफ्नो विकास चरण, जिम्मेवारी र क्षमताभन्दा पर जाने दायित्वहरू लिन आग्रह गरे पनि उनीहरूले गर्नुहुँदैन” भनेर थप्नुभयो। |
| Whisper Large v3 | उहाले उनिहरले अपनो बिकास चरन जिमेबारी र छमता बन्दा पर जाने दैत्तुहरू लिना आगरह गरे पनी उनिहरू गर्नु हुदैना बनेर थपनु भयो। |
Language detection
All 10 clips we sent without a language field were identified as Nepali. Auto-detection is safe here; sending language=ne still skips the detection step.
What to do instead
We would rather you know this before you build than after. For Nepali, 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 Nepali from Hugging Face run on your own hardware. If your audio mixes Nepali 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 Nepali?
On 30 clean read-speech clips it made 92% word errors. Most words come out wrong. Do not build on it for this language. Real recordings with noise, accents or several speakers will score worse.
Does language auto-detection work for Nepali?
In our test, yes: all 10 clips sent without a language were identified as Nepali. Sending language=ne still saves the detection step on every request.
Can it translate Nepali speech to English?
The API returns Nepali 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.