MultiTrack Add Subtitle To Video

You've spent a whole workflow generating a talking video, run speech-to-text, and now you want the words burned onto the frames. That's the whole job of MultiTrack Add Subtitle To Video - it takes your video, reads every subtitle segment sitting on the subtitle track of the MultiTrack editor, burns them in, and writes the matching SRT file while it's at it.

It's a finishing node, not a creative one, and that's fine. Most of the pack's subtitle path funnels into it: you recognize speech with Recognize Subtitle, the segments land on the multi-track's subtitle track, and this node is the terminal step that turns the whole timeline's subtitle data into an actual rendered video.

How it works

The node needs two things: the TRACKS_INFO output from a MultiTrack editor (that's where the subtitle track segments live) and the VIDEO you want them burned into. It collects every subtitle segment across the track, normalizes the timestamps against the video's own timing, and burns the text in - the same FFmpeg-style subtitle rendering the pack uses everywhere else in its video tools. So a subtitle track with five segments becomes five on-screen captions, no manual placement.

There's exactly one setting that matters, srt_save, with two options:

• temp (default) - the SRT goes to ComfyUI's temp folder. Perfect for a quick pass where you don't care about keeping the file.
• output - the SRT is written to output/srt, so you get a durable subtitle file alongside the video.

The single output is a VIDEO object, so you can wire the subtitled result straight into a Save Video node (or the pack's own easy saveVideo) rather than treating this node as the end of the line.

The workflow shape

A realistic pipeline: MultiTrack editor → TRACKS_INFO into Recognize Subtitle → subtitle text back onto the subtitle track → this node takes the track info plus the generated video → Save Video. If you're making dubs or interviews with a MiniMax H3 or LTX talking-head setup, this is the step that makes the result watchable without a separate editing app.

Installing the pack

MultiTrack Add Subtitle To Video ships inside ComfyUI-Easy-Media, so you install the whole pack:

cd ComfyUI/custom_nodes
git clone https://github.com/yolain/ComfyUI-Easy-Media.git

Then restart ComfyUI. You can also search "Easy-Media" in ComfyUI Manager. Two things to get right up front: FFmpeg must be installed on your system (the pack says so in big letters, and it's true - all the video/audio work sits on it), and subtitle recognition needs an ASR model. Whisper Large V3 goes in models/audio_encoders/, or Qwen3-ASR plus its forced aligner in models/Qwen3-ASR/ - some of these can be pulled automatically through the pack's built-in model download interface in the UI.

Where people get burned

The most common failure isn't this node - it's forgetting the ASR model entirely, in which case the subtitle track is empty and you burn nothing. Check that Recognize Subtitle actually returned text before you blame the burn-in. Also: burning subtitles re-encodes the video, so a short clip through this node takes a bit longer than a straight save, and a subtitle whose timestamps run past the video's duration will just get clipped. Nothing dramatic, just trim the last segment in the editor.

If you only ever want the SRT and not the burned video, grab the text from Recognize Subtitle directly instead - this node is for the finished product.