Audio or video to text. Drop a recording, or record one right here. Built for meetings, interviews, podcasts, and voice memos. Bring your own key. Use our server or your own.
Using shared free server
good for small files · slow for files over 100 MB
What kind of recording?
Language
Auto-detect works fine for most cases. Set it explicitly for non-English audio or better accuracy.
Tap Get Key above — opens AssemblyAI's signup in a new tab
Sign up with email or Google — no credit card required
After signup you land on your dashboard. Your API key is shown right at the top — tap to copy it
Paste it here and hit Save Key
Free tier includes $50 in credits — roughly 150+ hours of multi-speaker transcription. After that, AssemblyAI bills you directly (pay-as-you-go, no monthly fee).
✓
Saved in this browser only. Wiped from server memory after each request. Note: AssemblyAI processes the audio on their cloud to do speaker separation. Their data retention policy applies.
or
Recording
0:00
Your recording
0:000:00
Speed
File
How would you like to process this file?
⚡
Quick
Shared free server
No setup. Drop your file and we'll handle it on our shared server.
Estimated time: ~5 min
🔧
Advanced
Your own private server
Set up your own server once (5 min). After that, large files finish in a fraction of the time.
One-time setup: ~5 min · Per file after: ~5 min
0%
Set up your own server
Large files take a long time on our shared server. You can set up your own private server in about 5 minutes — you only do this once. After that, every transcription on this page will run on your own server, fast.
What you'll need
A free account at Render.com — the company that will host your server. Signing up takes about 30 seconds (you can use Google or GitHub).
About cost
Render will bill you directly, not us. Their free tier costs nothing and works fine for occasional use. If you transcribe often, their paid tier is $7 per month for much faster speed. You can choose either tier when you sign up — or upgrade later.
What will happen
We'll give you a short code, like purple-falcon. Keep this tab open.
You'll click a button. Render opens in a new tab.
You sign in to Render, then paste the code into the one field it asks for.
Click "Create" in Render. Walk away for 3–5 minutes while Render builds your server.
Come back to this tab — it will already be connected to your new server.
What you control
Your server is yours alone. Your audio files and your OpenAI key travel from your browser directly to your server — never through us. You can delete the server anytime from your Render dashboard.
Your unique pairing code:
Pairing Code — paste this in Render
purple-falcon
Now: click the button below to open Render in a new tab. Sign in if needed, then look for the field labeled PAIRING_CODE and paste your code there. Click "Apply" and "Create Web Service".
After clicking Create: come back to this tab. We'll watch for your server and connect automatically when it's ready (about 3–5 minutes).
Waiting for your server to come online…Usually 3–5 minutes after you click Create in Render
✓ A server appeared with your code
Confirm this is your server:
Host:
Only confirm if you just deployed this. If the URL looks unfamiliar, cancel and start over with a new code.
Already have your server URL? Paste it here
✓
You're connected!
Your file will now be transcribed using your private server. From now on, every transcription on this page will use your server until you change it back in the badge at the top.