$0.006/min · $0.36/hr
What the Whisper API actually costs.
OpenAI prices Whisper at $0.006/min. That's $0.36 per audio-hour — a number that stays invisible until the month you send a few hundred hours.
An audio-hour, priced by 5 providers
| Provider | $ / hr | Notes |
|---|---|---|
| AssemblyAI Universal-2 | $0.15 pay as you go | $0.15/hr async. Speaker diarization adds $0.02/hr. |
| Google Cloud Speech Dynamic batch (Standard) | $0.18 pay as you go | $0.003/min on the v2 API — batch jobs only, not real time. |
| OpenAI gpt-4o-mini-transcribe | $0.18 pay as you go | $0.003/min. OpenAI's cheapest transcription model. |
| SpeakEasy includes TTS Pro plan | $0.20 $10/mo ÷ 50 hrs | $10/month with 50 hours included, then $0.25/hr. Transcription and text-to-speech share the allowance — all 50 hrs spent on TTS is ~3.3M characters. |
| AssemblyAI Universal-3.5 Pro | $0.21 pay as you go | $0.21/hr async across 18 languages, diarization included. |
| Deepgram Nova-3 (pre-recorded) | $0.26 pay as you go | $0.0043/min monolingual; $0.0052/min ($0.31/hr) multilingual. Nova-2 is legacy-only. |
| OpenAI whisper-1 / gpt-4o-transcribe | $0.36 pay as you go | $0.006/min — the headline Whisper price, not OpenAI's floor. |
| Google Cloud Speech Standard recognition (v2) | $0.96 pay as you go | $0.016/min for the first 500k minutes each month. |
$0.20/hr is a plan rate, not a meter.
SpeakEasy is $10/month with 50 hours included, then $0.25 per additional hour. $0.20/hr is what you pay if you use all 50 hours; send only 10 hours and the same $10 works out to $1.00/hr. Every other provider here is true pay-as-you-go with no monthly minimum, so for transcription alone several of them cost less than this plan — do the arithmetic for your own volume before switching.
The 50 hours are a single shared allowance: transcription and text-to-speech both draw on it, so spending all of it on speech is ~3.3M characters — on the same OpenAI-compatible API, at no extra charge. None of the providers above bundle TTS with transcription at all.
Competitor prices read off each provider's own public pricing page on 2026-08-06; every provider name links to the page we read. Cheapest current pay-as-you-go rate shown for each, alongside their flagship model where it costs more.
What that becomes per month
| Volume | OpenAI whisper-1 | SpeakEasy plan |
|---|---|---|
| 10 hours / month | $3.60 | $10.00 |
| 50 hours / month | $18.00 | $10.00 |
| 200 hours / month | $72.00 | $47.50 |
| 1,000 hours / month | $360.00 | $247.50 |
SpeakEasy is $10/month with 50 hours included, then $0.25 per additional hour. Below 28hours a month, OpenAI's pay-as-you-go is genuinely cheaper — the plan only pays off once you're using it. And the OpenAI column here is whisper-1; their own gpt-4o-mini-transcribe at $0.18/hr costs less than this plan at every volume on this table.
Switching is one line
The API speaks the OpenAI protocol, so the official SDK works unchanged. Point base_url somewhere else and keep the rest of your code.
from openai import OpenAI
client = OpenAI(
api_key="sk-se-YOUR_KEY_HERE",
base_url="https://www.tryspeakeasy.io/api/v1", # the only line that changes
)
with open("audio.mp3", "rb") as f:
result = client.audio.transcriptions.create(
model="whisper-large-v3",
file=f,
)
print(result.text)
Get an API key → $1 for the first month, 50 hours included. Full Python walkthrough →
Test the output before you spend anything
This is the same whisper-large-v3 an API key would call. Drop a file you actually care about — the cost of that transcription is printed underneath the result.
FAQ
How much does the OpenAI Whisper API cost?+
$0.006/min of audio, billed by the second, which works out to $0.36 per audio-hour. There's no monthly minimum and no included tier — you pay for exactly what you send. New accounts get a small one-off credit, after which every minute is billed.
Is Whisper cheaper if I self-host it?+
Per hour of audio, often yes; per hour of your time, usually not. large-v3 wants roughly 10 GB of VRAM to run comfortably, so you're renting a GPU. That only beats $0.36/hr if the GPU stays busy — an idle instance you pay for by the hour loses to per-minute billing quickly. Self-hosting wins on steady high volume or when audio can't leave your network.
What is the cheapest Whisper API?+
Depends what you're comparing. Of the rows above that actually serve Whisper, OpenAI lists $0.36/hr and the SpeakEasy plan works out to $0.20/hr — but only if you fill all 50 included hours, because it's $10 a month either way. Send 10 hours and that same $10 is $1.00/hr. If you don't need Whisper specifically, cheaper models exist: AssemblyAI Universal-2 at $0.15/hr is the lowest rate in the table, and even OpenAI gpt-4o-mini-transcribe at $0.18/hr undercuts the plan at every volume. Every price links to the page we read it from on 2026-08-06.
Do I have to rewrite my code to switch?+
No. The API is OpenAI-compatible, so the official Python and Node SDKs work unchanged — you point base_url at a different host and keep everything else. That's the whole migration, and it's one line.
What counts as billable time?+
Audio duration, not processing time or file size. A 12-minute recording costs the same whether it's a 3 MB mono MP3 or a 200 MB WAV, and whether it returns in four seconds or forty.
Can I try it before paying anything?+
The widget on this page runs the same whisper-large-v3 model, free and without an account, five files a day. If the output isn't good enough for your use case, you'll know before you spend anything.