Repolode
Submit
murtaza-nasir

murtaza-nasir/speakr

Self-hosted web app that turns audio recordings into searchable notes with transcripts, speaker labels, summaries and chat.

Open-source alternative to:Otter.aiNotta

View on GitHub
Stars
4,159
Forks
352
Open issues
15
Last commit
2 days ago

What is it?

Speakr is a self-hosted web application for transcribing audio and organizing the results into searchable notes. You can record from a microphone or system and browser-tab audio, or upload existing files, and each recording gets a transcript with timestamps that play back in sync with the audio.

The transcription engine is pluggable. It can use a self-hosted WhisperX container (which enables speaker diarization and voice profiles), OpenAI, Mistral Voxtral, AssemblyAI, or other ASR services, and the connector is picked from your configuration. Summaries, action items, and per-recording chat are generated with an LLM through OpenRouter, OpenAI, or Ollama.

For libraries of many recordings, it offers folders, tags with their own prompts, retention policies, multi-user accounts with OIDC single sign-on, group sharing, a REST API, and signed webhooks. Recordings can be exported to templated files, such as Obsidian notes.

Key features

  • 01Records from microphone, system or browser-tab audio, or imports uploaded files.
  • 02Transcripts with timestamps that sync with playback.
  • 03Supports WhisperX, OpenAI, Mistral, AssemblyAI and other transcription backends.
  • 04Speaker diarization and voice profiles with the WhisperX backend.
  • 05Customizable summaries and tag-based prompts.
  • 06Semantic search and chat across the whole library.

Use cases

  • Keeping private meeting or interview recordings on your own server.
  • Turning lectures or voice memos into organized study notes.
  • Exporting transcripts automatically into an Obsidian or wiki vault.
  • Sharing recordings and summaries with a team through groups and permissions.

Quick start

  1. 1

    Create a project directory

    mkdir speakr && cd speakr
  2. 2

    Download the docker-compose configuration

    wget https://raw.githubusercontent.com/murtaza-nasir/speakr/master/config/docker-compose.example.yml -O docker-compose.yml
  3. 3

    Download the environment template and set your API keys

    wget https://raw.githubusercontent.com/murtaza-nasir/speakr/master/config/env.transcription.example -O .env
  4. 4

    Start Speakr with Docker Compose, then open http://localhost:8899

    docker compose up -d

Things to know

  • Needs API keys for transcription and summaries, or a GPU container for self-hosted WhisperX.
  • Version is labeled alpha (0.10.15-alpha), so breaking changes are possible.
  • Licensed under AGPL-3.0, a copyleft license that requires sharing modified source when offered as a network service.
  • Full features (speaker voice profiles, best diarization) require a GPU-backed WhisperX service.

In collections

Self-hosted appsAI video & audio tools

Topics

#transcription#speech-to-text#self-hosted#whisper#speaker-diarization#note-taking#llm#flask

FAQ

What is speakr?
Self-hosted web app that turns audio recordings into searchable notes with transcripts, speaker labels, summaries and chat. Speakr is a self-hosted web application for transcribing audio and organizing the results into searchable notes. You can record from a microphone or system and browser-tab audio, or upload existing files, and each recording gets a transcript with timestamps that play back in sync with the audio.
Is speakr free to use? What license does it have?
speakr is open source under the AGPL-3.0 license. Check the license terms before using it in commercial products.
Is speakr still maintained?
Actively maintained. The last commit to speakr was on Nov 5, 2026.
What can I use speakr for?
Keeping private meeting or interview recordings on your own server. Turning lectures or voice memos into organized study notes. Exporting transcripts automatically into an Obsidian or wiki vault. Sharing recordings and summaries with a team through groups and permissions.
How do I install speakr?
Create a project directory: mkdir speakr && cd speakr Download the docker-compose configuration: wget https://raw.githubusercontent.com/murtaza-nasir/speakr/master/config/docker-compose.example.yml -O docker-compose.yml Download the environment template and set your API keys: wget https://raw.githubusercontent.com/murtaza-nasir/speakr/master/config/env.transcription.example -O .env Start Speakr with Docker Compose, then open http://localhost:8899: docker compose up -d
Where can I get speakr?
The source code and installation instructions are on GitHub: https://github.com/murtaza-nasir/speakr

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

Related projects