ComfyUI-Stream-Pack

AudioTranscriptionNode

Real-time audio transcription node that buffers audio and outputs transcribed text on a controlled schedule to prevent message flooding.

Audio Utils · 1 input · 6 parameters · 1 output

FaceMeshDrawNode

A node that draws MediaPipe face mesh on images. Takes face mesh data from FaceMeshDetectNode and visualizes it. Optimized for visualization only, requires an input image. Uses the same drawing methodology as ControlNet Aux for optimal performance.

StreamPack/Preprocessors/Controlnet · 2 inputs · 3 parameters · 1 output

FaceMeshMaskNode

Creates binary masks for different regions of the face detected by FaceMeshDetectNode.

StreamPack/Preprocessors/Controlnet · 1 output

FaceMeshNode

A node that uses MediaPipe to detect face mesh from images and draws it. Optimized for real-time video processing. Uses the same drawing methodology as ControlNet Aux for optimal performance.

StreamPack/Preprocessors/Controlnet · 1 input · 9 parameters · 2 outputs

FeatureBankAttentionProcessor

Attention processor with feature injection.

SRTGeneratorNode

Generate SRT subtitle files from transcription with timing information. Compatible with AudioTranscriptionNode output when word timestamps are enabled.

Text Utils · 0 inputs · 5 parameters · 1 output

SuperResolutionModelLoader

Node that loads a super resolution model and outputs a reference to it

StreamPack/SuperResolution · 0 inputs · 2 parameters · 1 output

SuperResolutionUpscale

Node that applies a super resolution model to an image

StreamPack/SuperResolution · 2 inputs · 1 parameter · 1 output