Media Transcript

Play video/audio inside Obsidian with a synchronized, clickable transcript read from a same-named subtitle file next to the media. Click any line to seek; the playing line auto-highlights and scrolls into view.

This is a transcript player, not a transcriber — it does not do speech-to-text. Transcripts are produced by external tools; drop a .json / .srt / .vtt next to your media and it just works.

Features

• Opens .mp4 / .m4a / .mp3 / … and automatically loads a same-named subtitle in the same folder (.json / .srt / .vtt). If none exists, it shows a hint and does nothing else.
• Video: side-by-side layout — a large player on the left, transcript on the right, with a draggable divider (the ratio is remembered).
• Audio: a slim top control bar (title + player + speed) with a full-width transcript.
• Playback speed 0.75×–2× (video and audio).
• Click to seek, auto-highlight: click a line to seek and play; the current line highlights.
• Auto-scroll keeps the playing line centered — not stuck at the bottom edge. Scrolling by hand pauses it for a few seconds so you can read ahead; click a line to resume immediately.
• Adjustable transcript text size — A− / A+ in the transcript toolbar, or the slider in settings (10–32 px, remembered).
• Search the transcript — type in the search box to highlight every match; n/m counter with ↑ / ↓ (or Enter / Shift+Enter) to step through them, Esc to clear. Searching jumps forward from the line currently playing, and works for CJK text. The Search transcript command focuses the box, so you can bind a hotkey to it.
• Listen to a video without the picture — the 🎧 Audio only button swaps the video for a slim audio bar and a full-width transcript, keeping playback position and speed; 🎬 Video brings the picture back. Remembered across files and sessions.
• Right-click menu: play from here / copy timestamp / copy text; click a timestamp to copy it.
• Per-speaker labels (colored S0 / S1 …) when a JSON transcript has 2+ speakers.
• Export the current transcript to a Markdown note.

Installation

Via BRAT

1. Install the BRAT plugin.
2. Command palette → BRAT: Add a beta plugin for testing.
3. Enter semsevens/obsidian-media-transcript.
4. Enable Media Transcript in Community plugins.

Manual

Download main.js, manifest.json, and styles.css from the latest release into <vault>/.obsidian/plugins/media-transcript/, then enable the plugin.

From source

npm install
npm run build
cp main.js manifest.json styles.css <vault>/.obsidian/plugins/media-transcript/

Subtitle format

Place <media-name>.<marker>.{json,srt,vtt} next to the media, e.g. lecture.mp4 → lecture.whisper.json. JSON follows the Whisper verbose_json shape:

{ "segments": [ { "start": 0.0, "end": 4.2, "text": "…", "speaker": 0 } ] }

Settings

• Subtitle folder — leave empty to use each media file's own folder.
• Subtitle priority — order of filename [marker] tags when several subtitles exist.
• Video pane width — how much width the player takes in video mode (dragging the divider updates it too).
• Transcript font size — text size in px; the toolbar's A− / A+ change the same value.
• Auto-scroll transcript — keep the playing line centered while playing (on by default).
• Play videos as audio only — same as the toolbar's 🎧 button, as a default for new views.

If a media extension is already handled by Obsidian core or another plugin, use the Open in Media Transcript file-menu item or the Open current media file in transcript view command to open it in this view.

License

MIT