Urdu speech-to-text
FairWe sent 30 Urdu test clips through this API. Whisper Large v3 made 21% word errors, about one word in 5. $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 | 21% | 7.9% | 1 of 30 | 0 of 30 | 0.075× 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 | ان کے دور میں خفیف تلوُّث ایزا مسئلہ نہیں تھا جیسا کہ وہ آج ہے۔ ان کی سکونت ایسے شہروں یا حوزات میں ہے جہا ں تک بآسانی پہنچا جا سکتا ہے بہ نسبت ان مقامات کے جو ماڈرن زمانے میں تعمیر کئے گئے۔ |
| Whisper Large v3 | ان کے دور میں خفیق تلویس ایسا مسئلہ نہیں تھا جیسا کہ وہ آج ہے ان کی سکونت ایسے شہروں یا حوزات میں ہے جہاں تک بہت سانی پہنچا جا سکتا ہے بنسبت ان مقامات کے جو مادرن زمانے میں تعمیر کیے گئے |
Language detection
Send language=ur. Of 10 clips sent without it, only 5 were identified as Urdu; the others were labelled Hindi (5). When detection is wrong, Whisper writes the transcript in the language it guessed.
Quickstart
The request is the OpenAI transcription request with a different base URL. language is optional in the API but, as shown above, not optional in practice for Urdu.
curl https://api.inferenceapis.com/v1/audio/transcriptions \
-H "Authorization: Bearer $INFERENCE_API_KEY" \
-F "model=openai/whisper-large-v3" \
-F "language=ur" \
-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="ur", 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 Urdu test.
FAQ
How accurate is Whisper for Urdu?
On 30 clean read-speech clips it made 21% word errors, about one word in 5. 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 Urdu?
Not reliably. Of 10 clips sent without a language, 5 were identified as Urdu; the rest were labelled Hindi (5). A wrong label means the transcript is written in the wrong language, so always send language=ur.
Can it translate Urdu speech to English?
The API returns Urdu text. For English, transcribe first and translate the text with a chat model on the same key.
Related languages: Hindi (43%) · Punjabi (82%) · Sindhi (over 100%) · Persian (28%)