Dutch speech-to-text
ExcellentWe sent 30 Dutch test clips through this API. Whisper Large v3 made 7.1% word errors, about one word in 14; Parakeet made 7.8%. $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 | 7.1% | 1.6% | 11 of 30 | 0 of 30 | 0.062× real time |
Parakeet TDT 0.6B v3nvidia/parakeet-tdt-0.6b-v3 | 7.8% | 2.8% | 6 of 30 | 0 of 30 | 0.047× real time |
Excellent. Production quality. Transcripts can be used as they are for notes, search and subtitles. 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 | Vorige week kondigde METI aan dat Apple 34 aanvullende incidenten met oververhitting had gemeld die het bedrijf 'niet ernstig' noemde. |
| Whisper Large v3 | Vorige week kondigde Meti aan dat Apple 34 aanvullende incidenten met overberitting had gemeld die het bedrijf niet ernstig noemde. |
| Parakeet (its median clip) | Spoken: Varkens dragen de ziekte bij zich en dragen het via muggen over op mensen. Vakers dragen de ziekte bij zich en dragen het via muggen over op mensen. |
Language detection
All 10 clips we sent without a language field were identified as Dutch. Auto-detection is safe here; sending language=nl still skips the detection step.
Which model for Dutch
The two models are within the noise of a 30-clip test (7.1% and 7.8%). Parakeet is faster; Whisper handles mixed-language audio and has automatic detection across 100 languages. Both cost the same, so try both on a sample of your own recordings. 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=nl" \
-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="nl", 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 Dutch test.
FAQ
How accurate is Whisper for Dutch?
On 30 clean read-speech clips it made 7.1% word errors, about one word in 14. Production quality. Transcripts can be used as they are for notes, search and subtitles. Real recordings with noise, accents or several speakers will score worse.
Does language auto-detection work for Dutch?
In our test, yes: all 10 clips sent without a language were identified as Dutch. Sending language=nl still saves the detection step on every request.
Whisper or Parakeet for Dutch?
On the same clips Whisper Large v3 scored 7.1% and Parakeet 7.8%. That is within the noise of a 30-clip test; Parakeet is faster, so try it first on long recordings. Both cost $0.002 per minute.
Can it translate Dutch speech to English?
The API returns Dutch text. For English, transcribe first and translate the text with a chat model on the same key.