Hear the difference
People talk about three times faster than they type, but speech comes with “um” and “uh”. Pick a sample and press play: on the left is what was said, on the right what SparkFlow gives you to paste.
How it works
Speak
Click the mic and talk naturally. There’s nothing to install and nothing to set up.
Capture
Text appears as you talk, with filler words removed and sentences capitalized and punctuated.
Use it anywhere
Copy it with one click and paste it into an AI chat or coding agent, an email, a message or your notes.
Keep your ideas flowing
Clean text, right away
Filler words like “um”, “uh” and “hmm” disappear, and every sentence starts with a capital and ends with punctuation.
Sorted into topics
Every transcript is saved and sorted into topics for you. Make your own topics with keywords and colors.
Activity rings
See how much typing time you’ve saved each day, week and month, against goals you set.
Your history, by day
Browse past thoughts on a calendar, grouped by topic or time, and edit, re-topic or delete them in bulk.
Your transcripts stay in your browser
No accounts, no analytics and no SparkFlow server. Your transcripts, topics and history are saved in your own browser and nowhere else. The speech recognition itself is done by your browser; in Chrome and Edge, that means the browser maker’s speech service hears the audio.
Built in the open
SparkFlow is free and open source, and help of every size is welcome: try it in your work and say what felt slow, suggest an idea, or send a change. Language choices, voice commands like “new line”, and exporting your history are next on the wish list.