Diktafon icon
Diktafon icon

Diktafon

 20 likes

Private, open-source app records memos with on-device transcription, summary, topic-based cassette organization, auto language detection, background recording, editable transcripts, and portable export formats—no accounts, cloud, analytics, or ads.

Diktafon screenshot 1
Diktafon screenshot 2
+4
Diktafon screenshot 3

Cost / License

  • Free
  • Open Source

Application type

Platforms

  • Linux
  • Android
  • iPhone
  • F-Droid
Diktafon alternatives  
20likes
19alternatives
0articles
Save

Features

Properties

Features

 Tags

Highlights All activities

Diktafon News & Activities

Recent activities

Ad
Screenshot
start.me icon
start.me

Manages bookmarks and RSS feeds with customizable widgets and themes. Offers browser extensions for cross-device usability and personalized pages.

 
Learn More

Diktafon information

AlternativeTo Categories

AI Tools & Services, Office & Productivity, Audio & Music

Apple AppStore

  •   Updated 
  •   4.5 avg rating
View in AppStore

GitHub repository

  •  278 Stars
  •  21 Forks
  •  31 Open Issues
  •   Updated  
View on GitHub

Popular alternatives

View all
Diktafon was added to AlternativeTo by jaromiru on and this page was last updated .

No comments or reviews, maybe you want to be first?

Official Links

AppStores & Other Links

Download from Google Play Store
Download from iPhone App Store

What is Diktafon?

Diktafon is a voice-memo app where memos live on topic-based cassettes, and each cassette plays as one continuous tape of chronologically ordered recordings.

Every memo is transcribed and every cassette summarised and titled — fully on-device (Whisper for transcription, Qwen for summaries). No accounts, no cloud, no analytics, no ads. Tap any word in a transcript and the tape jumps there; transcripts can also be edited by hand. Recording carries on with the screen off or the app in the background. Spoken language is auto-detected per memo, and the app itself speaks 20 languages. Cassettes export as portable archives with the audio and readable text inside. Free and open source (MIT).

The speech and summary models are downloaded once on first run (Wi-Fi recommended); 4 GB RAM is recommended for the default models.