A free speech to text tool that turns your voice into text, live from your microphone or from a recording, on your own device.
Talk into your microphone, or upload a recording, and Whisper writes it out on your device. Nothing is uploaded.
Add your recording
Private: nothing leaves your device. The model downloads once (77 MB) and your browser keeps it.
This free speech to text tool turns your voice into text: talk into your microphone, or upload a recording.
Whisper runs in your browser, so what you say never leaves your computer. Every line keeps its time.
Here's how to dictate with it, how to transcribe a file, and how to get the cleanest text.
What is speech to text?
Speech to text turns spoken words into written text. This tool does it two ways: live from your microphone on the Talk tab, or from a recording you upload.
It runs Whisper, which OpenAI describes as "a general-purpose speech recognition model."
Talk into the page
The Talk tab records you in the browser, with a live level meter. When you press Stop and transcribe, the take is written out on your device.
Your browser asks to use the microphone first. Nothing is recorded until you press Start talking, and nothing is sent anywhere.
On your device or on our server
The page opens on your device. Whisper, OpenAI's speech recognition model, runs in your browser, and the file never leaves it.
For long files, pick On our server. The sound goes in 2-minute pieces to Groq's Whisper and comes back as text in seconds.
| On this device | On our server | |
|---|---|---|
| Privacy | Nothing leaves your device | Sent in pieces, not kept |
| Download | 77 MB once, or 563 MB for Best | None |
| Length | No limit | Up to 30 minutes a file |
| Cost | Free | Free, within a daily limit |
Groq, the company that runs the server's Whisper, says in its data policy that "by default, Groq does not retain customer data for inference requests." We don't keep the sound either.
Standard or Best
Standard is whisper-base, a 77 MB download that runs in any modern browser. On a 2021 MacBook Pro, it wrote a 23-second clip in 5 seconds.
Best is whisper-large-v3-turbo, 563 MB, on browsers with WebGPU. It wrote 100 seconds in 10, and a Spanish clip word for word, after its download.
| Model | Download | Runs on | Best for |
|---|---|---|---|
| Standard | 77 MB | Any modern browser | Quick notes, clear speech |
| Best | 563 MB | Browsers with WebGPU | Names, accents, other languages |
Without WebGPU, Best is whisper-small, 249 MB, which is slower. Each model downloads once, and your browser keeps it.
Languages
Whisper was trained on "680,000 hours of multilingual and multitask supervision," says its paper. The language list has every language it knows, from Afrikaans to Yoruba.
Leave it on Detect it, and the tool hears the language in the first words and keeps it for the rest. Pick the language when you know it: names and short clips come out better.
Timestamps, SRT and VTT
Every line keeps its time. Press a time and the recording plays from there, and the line being said lights up as it plays.
Download it as plain text, Markdown, SRT or VTT. SRT and VTT are the caption files YouTube, Vimeo and video editors take.
Get a cleaner transcript
- Pick the language – when you know it, instead of Detect it.
- Use Best – for names, accents and anything but clear English.
- Record close to the mic – your voice rises above the room's noise.
Two things cost the most words:
- Music under the voice – Whisper can drop words, or invent a line in a long music break.
- People talking over each other – Whisper follows one voice at a time.
After that, Add punctuation in step 3 tidies the text in your own AI, without changing a word.
After you dictate
Ask your own AI to tidy it, turn it into a post, or make it a to-do list, in step 3. To record a longer take with markers and trims, use the audio recorder.
Who is the speech to text for?
It's for anyone who talks faster than they type:
- Audio recorder – people who record first and transcribe after
- Voice memo to text – anyone with ideas saved as voice memos
- Spanish transcription – Spanish speakers
- Audio translator – people who need the words in another language
Every other tool is on my free tools page.
Do more with your recording
Dictate on your Mac with
Speech to text FAQs
Questions about the speech to text? Here's what to know.
Speech to text turns spoken words into written text.
This tool does it from your microphone or from a recording, with Whisper in your browser.
Open the Talk tab, press Start talking, then Stop and transcribe.
Your take is written out on your device.
Yes, in a modern phone browser, with the microphone or a recording.
Long files are quicker On our server on a phone.
No, on this device the recording and the text stay in your browser.
Only the model is downloaded, once.
Yes, this speech to text tool is free, with no account and no limit on your device.
On our server, each visitor gets 40 minutes a day.
No, not on this device: Whisper runs in your browser and the file never leaves it.
On our server, the sound is sent in pieces, written out, and not kept.
A translation sends the text only.
No, Whisper writes what's said without names.
Add the names yourself in the text, or in your editor.
This speech to text tool is for anyone who'd rather talk than type:
- Podcast transcript generator – podcasters who need show notes and quotes.
- Audio converter – anyone who records and edits audio.
- Audio recorder – people who record interviews and notes in the browser.
Writers use it to draft out loud, and people who find typing hard use it to write anything.
I did.
I'm Navid Moazzez, and the speech to text tool is one of my free tools on navid.me.
Read more about me.
There's an audio to text converter, a video to text converter, speech to text, a voice memo to text converter, Spanish transcription, an audio translator and a podcast transcript generator.
The rest are in free tools.
Navid Moazzez
AI business strategist & AI OS builder
Navid Moazzez helps creators and founders master AI and build their own AI Operating System (AI OS) to automate their business and life.
Navid.me is reader-supported. When you buy through links on this site, I may earn an affiliate commission. Learn more.
More free tools
Apple Podcasts
Spotify
X
Bluesky
LinkedIn
Threads
Mastodon
Instagram
TikTok
Pinterest
YouTube
Amazon KDP
Apple Books
Amazon
Free AI newsletterThe most actionable AI newsletter for founders
Every week, get proven AI strategies, curated tools, and step-by-step systems to grow your audience, create better content, and build a profitable creator business.
No fluff, no filler, no BS. Just five minutes each week that might level up your online business and life.
P.S. Sign up now to get free access to my ultimate AI tools guide for creators.
