🎤 Voice generation

Documents are created by voice.

Just start speaking. Rasepi uses speech-to-text to transform your voice into neatly structured documents. Headings, paragraphs, lists: everything is created as you speak. No typing, no blank page. And because Rasepi is multilingual, speak in your language and publish in 40+ others.

🎤 Speech Capture Live Transcription
“Heading: Onboarding for New Engineers”
Onboarding for New Engineers
“New Paragraph”
On your first day, we’ll set up your laptop and access credentials. Then the three-day Engineering Immersion Program begins.
“Bullet points: Provide laptop, set up access credentials, assign mentor”
Being transcribed & formatted…

The blank page is the greatest enemy of good documentation.

Most good content never gets created because writing it down is tedious. Knowledge resides in people’s heads. But there’s too much friction between a thought and a polished document.

❌ Traditional typing

  • The blank page blocks you before you even start
  • Formatting interrupts the flow of thought
  • Knowledge remains unwritten because there’s no time
  • Hardly possible on the go or freehand
  • Structure only emerges after laborious rework

✔ Rasepi by voice

  • Just get started—talk instead of staring
  • Rasepi automatically formats blocks as you speak
  • Knowledge is captured while it’s fresh
  • Capture hands-free, at your desk or on the go
  • Structure emerges in real time, not afterward

How it works

Speech-to-text meets Rasepi’s block-based architecture. Spoken language is transformed into clean, structured documents.

1

Activate the microphone and speak

Tap the microphone and speak normally. Rasepi transcribes your voice in real time directly into the editor. No uploading, no waiting.

2

Voice commands for structure

Say “Heading,” “New Paragraph,” or “Bulleted List,” and Rasepi formats the appropriate block. Punctuation and layout are generated as you speak.

3

Clean blocks instead of a wall of text

Each section becomes its own block with a unique ID. This is the same foundation that enables block-level translation and version tracking.

4

Refine and publish

Draft the rough version freely and polish it later at your leisure. With a single click, Rasepi translates into 40+ languages—only the changed blocks.

Language makes documentation accessible

Rasepi’s block-based architecture is why speech-to-text here is more than just a dictation tool. Spoken language isn’t dumped into a wall of text, but translated into clean, individually addressable blocks.

This means: Each dictated section receives its own block ID and thus the same benefits as typed content: block-level translation, independent version tracking, and versioning.

Speak once. Publish in any language.

Speak in German. Rasepi automatically translates into 40+ languages via DeepL and only re-transmits the changed blocks. How multilingual publishing works →

Key Features

🎤 Real-time transcription

Your voice appears immediately as text in the editor, no uploading, no waiting.

🧾 Voice commands

Headings, paragraphs, lists, and links via voice. Structure is created as you speak.

🌐 Speak in any language

Dictate in German, English, or any other language. Publish in 40+ languages.

🧰 Clean block structure

Each section becomes its own block with an ID, ready for translation and version tracking.

Speak. Rasepi takes notes.

Dictate entire documents by voice. Rasepi automatically formats the blocks and publishes in 40+ languages.

Join the waitlist → Contact us

Rasepi is in private beta. We’re inviting teams in waves.