Transcripts for less than a cent an hour.
Send a call, meeting or episode and get the words back, for $0.009 an hour of audio. The endpoint matches OpenAI’s, so existing code only needs a new base URL.
- $0.009 an hour of audio
- OpenAI SDK-compatible endpoint
- 0 bytes of your audio stored
How it works
- 01
Send the recording
WAV, MP3, FLAC or OGG, up to 25 MB and 10 minutes per request.
- 02
Set the language
English by default; pass a code like es, ja or pt for anything else.
- 03
Get the text
JSON with the transcript, language and duration, or plain text.
import os
from openai import OpenAI
client = OpenAI(base_url="https://api.iodine.dev/v1", api_key=os.environ["IODINE_API_KEY"])
with open("meeting.mp3", "rb") as f:
result = client.audio.transcriptions.create(model="whisper-1", file=f, language="en")
print(result.text)What you get
Drop-in for OpenAI
/v1/audio/transcriptions works with OpenAI’s SDKs. Change the base URL and key.
Learn more ›Never stored
Recordings are transcribed in memory; we keep only the length, language and cost.
Learn more ›Try it without code
Upload or record audio in the console’s Playground at the same price.
Learn more ›One balance
Transcription draws on the same prepaid credit as text to speech.
Learn more ›
What it costs
1,000 hours of recorded calls and meetings.
60,000 minutes on Iodine $9.00
The same at OpenAI Whisper’s published price $360.00
At Iodine’s pay-as-you-go prices. Other providers at their published pay-as-you-go rates; see the price comparison.
Questions
Up to 10 minutes and 25 MB per request. Split longer recordings into parts and join the text.
Pass the language spoken as a code like en, es, ja or pt; English is the default. Set it for anything that isn’t English.
Only the base URL and the key. Fields we don’t use, like model and prompt, are ignored.
Start building with Iodine.
Top up from $5. No plan, seats or minimums.