Repolode
Submit
rishikanthc

rishikanthc/Scriberr

Self-hosted, offline audio and video transcription app with speaker detection, chat with transcripts, and a folder watcher for automation.

Open-source alternative to:PlaudOtter.ai

View on GitHubWebsite
Stars
3,139
Forks
274
Open issues
109
Last commit
2 months ago

What is it?

Scriberr is a self-hosted application that transcribes audio and video files locally, so recordings are not sent to a third-party cloud service. It runs state-of-the-art models such as NVIDIA Parakeet and Canary as well as Whisper, producing text with word-level timing.

Beyond transcription, it detects and labels individual speakers (diarization), lets you chat with transcripts through Ollama or OpenAI-compatible providers to generate summaries or ask questions, and includes an in-app recorder and note-taking for annotating recordings. A folder watcher and an API let it plug into automation tools such as n8n.

The web interface can be installed as a PWA on desktop or mobile. It is distributed through Homebrew, Docker, and Docker Compose, with separate images for CPU and NVIDIA CUDA, including a Blackwell-specific build for RTX 50-series GPUs.

Key features

  • 01Offline transcription using Whisper, Parakeet, and Canary models with word-level timestamps
  • 02Automatic speaker detection and labeling (diarization)
  • 03Chat with transcripts and generate summaries via Ollama or OpenAI-compatible APIs
  • 04Folder watcher that automatically processes new files
  • 05Built-in audio recorder and notes for annotating transcripts
  • 06Installable PWA with light and dark mode

Use cases

  • Transcribing voice memos or recordings from a dictation device without cloud uploads
  • Turning long meetings or interviews into searchable transcripts with speaker labels
  • Automating transcription of files dropped into a folder through n8n workflows
  • Summarizing recordings with a locally hosted LLM

Quick start

  1. 1

    Add the Scriberr Homebrew tap

    brew tap rishikanthc/scriberr
  2. 2

    Install Scriberr with Homebrew (also installs UV)

    brew install scriberr
  3. 3

    Start the server, then open http://localhost:8080 in your browser

    scriberr

Things to know

  • The README notes that development was paused for a period and is only now resuming, so maintenance pace may be uneven.
  • Upgrading from v1.1.0 requires changing Docker volume mounts and deleting the old whisperx-env folder.
  • Serving over plain HTTP requires setting SECURE_COOKIES=false, or audio playback fails.

In collections

Self-hosted appsPrivacy & local-first toolsAI video & audio tools

Topics

#speech-to-text#whisper#transcription#self-hosted#diarization#offline#audio#privacy#ai#transcript

FAQ

What is Scriberr?
Self-hosted, offline audio and video transcription app with speaker detection, chat with transcripts, and a folder watcher for automation. Scriberr is a self-hosted application that transcribes audio and video files locally, so recordings are not sent to a third-party cloud service. It runs state-of-the-art models such as NVIDIA Parakeet and Canary as well as Whisper, producing text with word-level timing.
Is Scriberr free to use? What license does it have?
Scriberr is open source under the MIT license. Check the license terms before using it in commercial products.
Is Scriberr still maintained?
Updated within 6 months. The last commit to Scriberr was on Sep 20, 2026.
What can I use Scriberr for?
Transcribing voice memos or recordings from a dictation device without cloud uploads Turning long meetings or interviews into searchable transcripts with speaker labels Automating transcription of files dropped into a folder through n8n workflows Summarizing recordings with a locally hosted LLM
How do I install Scriberr?
Add the Scriberr Homebrew tap: brew tap rishikanthc/scriberr Install Scriberr with Homebrew (also installs UV): brew install scriberr Start the server, then open http://localhost:8080 in your browser: scriberr
Where can I get Scriberr?
The source code and installation instructions are on GitHub: https://github.com/rishikanthc/Scriberr

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects