Danish speech-to-text
GoodWe sent 30 Danish test clips through this API. Whisper Large v3 made 15% word errors, about one word in 7; Parakeet made 20%. $0.002 per audio minute, word timestamps, SRT and VTT subtitles, files up to 100 MB.
Measured accuracy
| Model | Word error rate | Character error rate | Clips with no errors | Empty transcripts | Speed |
|---|---|---|---|---|---|
Whisper Large v3openai/whisper-large-v3 | 15% | 5.6% | 0 of 30 | 0 of 30 | 0.062× real time |
Parakeet TDT 0.6B v3nvidia/parakeet-tdt-0.6b-v3 | 20% | 7.4% | 1 of 30 | 0 of 30 | 0.063× real time |
Good. Usable with light review. Expect to correct a word here and there, mostly names and numbers. 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 | Dette er en vigtig måde at skelne mellem nogle verber og objektiver |
| Whisper Large v3 | Det er en vigtig måde at skælde mellem nogle verber og objektiver. |
| Parakeet (its median clip) | Spoken: Andre brydere hyldede også Luna. Andre brødere hyldede også Luna. |
Language detection
All 10 clips we sent without a language field were identified as Danish. Auto-detection is safe here; sending language=da still skips the detection step.
Which model for Danish
Whisper Large v3 was more accurate on these clips (15% against 20%). Parakeet is faster, which matters mainly on long recordings. Both cost the same. Side by side.
Quickstart
The request is the OpenAI transcription request with a different base URL. language is optional.
curl https://api.inferenceapis.com/v1/audio/transcriptions \
-H "Authorization: Bearer $INFERENCE_API_KEY" \
-F "model=openai/whisper-large-v3" \
-F "language=da" \
-F "file=@recording.mp3" \
-F "response_format=verbose_json"import os
from openai import OpenAI
client = OpenAI(base_url="https://api.inferenceapis.com/v1", api_key=os.environ["INFERENCE_API_KEY"])
with open("recording.mp3", "rb") as f:
r = client.audio.transcriptions.create(model="openai/whisper-large-v3", file=f, language="da", response_format="verbose_json")
print(r.text)Subtitles and timestamps
response_format=srt or vtt returns a subtitle file with cues of up to two lines, broken at sentence ends and pauses. verbose_json with timestamp_granularities[]=word returns a start and end time for every word if you want to build your own.
Cost
$0.002 per minute of audio on either model: $0.12 for an hour, $12 for a hundred hours. Duration is measured from the file you upload. There is no daily cap on minutes and no subscription; enough to run your own Danish test.
FAQ
How accurate is Whisper for Danish?
On 30 clean read-speech clips it made 15% word errors, about one word in 7. Usable with light review. Expect to correct a word here and there, mostly names and numbers. Real recordings with noise, accents or several speakers will score worse.
Does language auto-detection work for Danish?
In our test, yes: all 10 clips sent without a language were identified as Danish. Sending language=da still saves the detection step on every request.
Whisper or Parakeet for Danish?
On the same clips Whisper Large v3 scored 15% and Parakeet 20%. Whisper was more accurate here; Parakeet is faster, which matters mainly for long recordings. Both cost $0.002 per minute.
Can it translate Danish speech to English?
The API returns Danish text. For English, transcribe first and translate the text with a chat model on the same key.