Flux TTS

approved

by Saunak Karnati

Record audio from the ribbon, save it to your vault, and create Groq-powered transcription notes. - This plugin has not been manually reviewed by Obsidian staff.

↓ 72 downloadsUpdated 20d agoMIT
Install in ObsidianView on GitHub

Flux TTS

An Obsidian plugin that records audio from the ribbon, saves it to your vault, transcribes it with Groq Whisper, and creates a note with the transcript and an audio embed.

Features

  • One-click recording from the ribbon, or command-palette commands (toggle/start/stop).
  • Groq Whisper transcription — whisper-large-v3-turbo (fast) or whisper-large-v3 (accurate), switchable in settings.
  • Inline transcription — transcribe straight into the note at your cursor, with a footnote linking back to the recording.
  • Timestamped segments (optional) — each line links to its moment in the audio.
  • AI cleanup (optional) — a Groq LLM fixes punctuation and removes filler words; the original transcript is kept, never discarded.
  • Live waveform while recording, so you can see your mic is working.
  • Media-link transcripts (opt in) — paste a YouTube or public media link to add its transcript as a citation-style footnote.
  • Note templates from a vault file, live filename previews, and validation that catches malformed {{placeholders}}.
  • Warns and auto-stops on a silent or disconnected mic. Works on desktop and mobile.

Setup

  1. Create a free API key at console.groq.com/keys.
  2. Open Settings → Flux TTS and paste the key (stored in Obsidian's secret storage, not plain text).
  3. Click the microphone ribbon icon to start; click again to stop. The note opens automatically.

The Settings tab covers the model, recording start delay, audio/note save locations, filename and note templates, AI cleanup, and timestamped segments — each with inline help.

Compatibility

Requires Obsidian 1.12.0+. The settings screen uses the stable imperative settings API available in Obsidian 1.12.

Privacy

Audio is saved locally. Transcription (and AI cleanup, if enabled) sends data to Groq using your API key. When media-link transcripts are enabled, pasted links are fetched to locate captions or public media; media without captions is sent to Groq for transcription. Nothing is sent until you record or enable and use media-link transcripts.

Roadmap & development

See ROADMAP.md for what's next. Source is TypeScript in src/, bundled to main.js with esbuild:

npm install
npm run dev    # watch mode
npm run build  # type-check + production build

Releases are built by GitHub Actions: pushing a tag matching the manifest.json version attaches main.js, manifest.json, and styles.css to a release.

License

MIT

For plugin developers

Search results and similarity scores are powered by semantic analysis of your plugin's README. If your plugin isn't appearing for searches you'd expect, try updating your README to clearly describe your plugin's purpose, features, and use cases.