Batch Audio Streams

Batch Audio Streams is the audio-flavored twin of Batch Video Streams in ComfyUI Core Video Nodes (Immac/ComfyUI-CoreVideoMocks). Its job in a real pipeline: gather multiple audio streams into one batch so you can pick one later. Multiple soundtracks in a single file are the norm in the real world - the director's commentary track on a disc, the original language plus a dub, a music-only mix alongside the full mix. A muxed video file can absolutely hold several audio streams, which is why the batch concept exists at all.

Two required inputs, both force-input STRINGs: audio_stream_1 and audio_stream_2. One output, batched_audio_streams, a STRING. In the pack's proposed API you'd feed the batch to Get Audio Stream with an index and pull out the track you want. The pair tells the same story as its video twin: batch first, select second.

What actually happens

This is a mock, and the output is a string that echoes your two inputs back: "Mock Batched Audio Streams: ... + ...". No real audio is touched, no batch object exists, nothing you can feed to a player. The author's own README is upfront that the whole pack is "a mock of a possible implementation" of core video nodes, and GitHub calls it "A set of mock nodes for RFC purposes." This node is a shape, not a tool.

Install

cd ComfyUI/custom_nodes
git clone https://github.com/Immac/ComfyUI-CoreVideoMocks

Restart ComfyUI. No pip packages, no model files, no ffmpeg - pure Python on ComfyUI's own folder_paths, also installable via ComfyUI Manager under "ComfyUI Core Video Nodes."

The bottom line

Useful only as a specimen. If you actually need to extract a specific audio track from a file, that's ffmpeg's -map and on the ComfyUI side VideoHelperSuite (VHS) is where real video/audio work happens. This pack sits at two GitHub stars with no community discussion, untouched since March 2025. Study the batch-then-select pattern if you're designing nodes; don't expect it to run anything.