Scriberr

Scriberr transcribes recordings locally for people who want to review meetings and voice notes without uploading audio to a cloud service. It combines transcription with playback, annotations and optional questions about the recorded conversation.

• Transcription and speakers: Whisper, NVIDIA Parakeet and Canary produce text with word-level timestamps, while automatic speaker detection labels different speakers.
• Recording review: The transcript follows playback, and selecting text jumps to that point in the recording; highlights and notes let you mark passages while listening.
• Transcript chat: Connect Ollama or an OpenAI-compatible provider to ask questions and generate summaries; a local model through Ollama keeps this processing on your machine, while a cloud provider receives transcript content.
• Capture and automation: A built-in recorder captures audio, and a folder watcher and API connect incoming recordings to automated workflows such as n8n.

Scriberr is free and open source under the MIT license. Transcription works offline on your hardware. Docker deployment supports CPU operation without a dedicated GPU or acceleration with a compatible NVIDIA GPU, including Blackwell hardware. Its browser interface can also be installed as a web app on desktop and mobile.