Persian speech-to-text
FairWe sent 30 Persian test clips through this API. Whisper Large v3 made 28% word errors, about one word in 4. $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 | 28% | 7.2% | 1 of 30 | 0 of 30 | 0.054× real time |
Fair. Gets the content across, but needs a human pass before it is published or quoted. 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 | این سه پادشاهی یکی از خونین‌ترین دوران تاریخ چین باستان به حساب می‌آید که در آن هزاران نفر برای تصاحب بالاترین کرسی در کاخ بزرگ در شی‌آن جان خود را از دست دادند. |
| Whisper Large v3 | این سپاتشاهی یکی از خونین ترین دوران تاریخ چین باستان به حساب می آید که در آن هزاران نفر برای تحاسب بلاترین کرسی در کاخ بزرگ در شیان جان خود را از دست دادن. |
Language detection
All 10 clips we sent without a language field were identified as Persian. Auto-detection is safe here; sending language=fa 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=fa" \
-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="fa", 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 Persian test.
FAQ
How accurate is Whisper for Persian?
On 30 clean read-speech clips it made 28% word errors, about one word in 4. Gets the content across, but needs a human pass before it is published or quoted. Real recordings with noise, accents or several speakers will score worse.
Does language auto-detection work for Persian?
In our test, yes: all 10 clips sent without a language were identified as Persian. Sending language=fa still saves the detection step on every request.
Can it translate Persian speech to English?
The API returns Persian text. For English, transcribe first and translate the text with a chat model on the same key.