Speakr

A self-hosted transcription app that turns recordings into searchable notes, with WhisperX speaker recognition and local or cloud transcription.

Alternative to:

Speakr is a self-hosted transcription and note-taking app for people who need to revisit meetings, interviews or lectures. It pairs recordings with searchable transcripts and AI summaries.

Audio capture: Record microphone and system or browser-tab audio together, upload existing files, or automatically import recordings from a watched folder.
Speakers and playback: Jump from a transcript line to the matching audio, label speakers, and use the WhisperX backend to recognize recurring voices across recordings.
Summaries and questions: Generate summaries with custom prompts, extract action items and events, ask questions about a recording with timestamp references, and search across the library through Inquire Mode.
Library and sharing: Organize recordings with folders and tags that carry processing prompts, share with people or groups, and control viewing, editing and resharing permissions.
Workflow connections: Use the REST API and signed webhooks to connect with n8n, Zapier or Make, and export transcripts as text or Markdown.

The open-source app uses AGPL-3.0 and runs on your server, with Docker or manual deployment and a browser interface. Transcription can use a self-hosted WhisperX service or cloud providers such as OpenAI, Mistral and AssemblyAI; choosing a cloud provider sends audio there for processing. Custom vocabulary helps with names and specialist terms, particularly with WhisperX. Retention controls can delete audio while preserving its transcript.