Korean speech-to-text
GoodWe sent 30 Korean test clips through this API. Whisper Large v3 made 13% word errors, about one word in 7. $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 | 13% | 2.2% | 9 of 30 | 0 of 30 | 0.061× 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 | 합금은 기본적으로 금속 두 개 이상의 혼합물이다. 주기율표 상에 원소가 많이 있다는 것을 잊지 마라. |
| Whisper Large v3 | 합금은 기본적으로 금속 2개 이상의 혼합물이다. 주기율표 상에 원소가 많이 있다는 것을 잊지 마라. |
Language detection
All 10 clips we sent without a language field were identified as Korean. Auto-detection is safe here; sending language=ko still skips the detection step.
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=ko" \
-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="ko", 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 Korean test.
FAQ
How accurate is Whisper for Korean?
On 30 clean read-speech clips it made 13% 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 Korean?
In our test, yes: all 10 clips sent without a language were identified as Korean. Sending language=ko still saves the detection step on every request.
Can it translate Korean speech to English?
The API returns Korean text. For English, transcribe first and translate the text with a chat model on the same key.