This plugin integrates OpenAI's Whisper speech-to-text model into Obsidian, allowing users to transcribe spoken words into written notes directly within their knowledge base. It targets Obsidian users who frequently take notes via voice or wish to convert existing audio files into text. The primary benefit is seamless voice-to-text functionality within the Obsidian environment, enhancing note-taking efficiency.

How It Works

The plugin leverages the OpenAI Whisper API for transcription. Users can record audio directly within Obsidian or upload existing audio files. The transcribed text is then saved as a new note or appended to the current cursor position, with options to save the original audio file. This approach offloads the computationally intensive transcription process to OpenAI's robust infrastructure.

Quick Start & Requirements

• Install via Obsidian's Community Plugins.
• Requires an OpenAI API key.
• Settings allow configuration of API URL, model, language, and save locations for transcriptions and audio.
• Manual installation instructions are provided.

Highlighted Details

• Supports both live recording and transcription of existing audio files.
• Configurable save paths for transcriptions and audio files within the Obsidian vault.
• Offers a keyboard shortcut (Alt + Q) for quick recording start/stop.
• Transcriptions can be saved as new notes or appended to the current cursor.

Maintenance & Community

The plugin is maintained by nikdanilov, who is active on Twitter (@nikdanilov_). Contributions and issue reports are welcomed via the GitHub Issues page.

Licensing & Compatibility

The plugin's license is not explicitly stated in the README. Compatibility with Obsidian's plugin ecosystem is implied by its availability through Community Plugins.

Limitations & Caveats

The plugin relies on an external OpenAI API key, incurring potential costs based on usage. The README does not specify which Whisper models are supported or if there are any limitations on audio file formats or lengths.