Cleanup the repo and add help popup descriptions

yvann-ba · Sep 7, 2024 · 3aacb91 · 3aacb91
1 parent c5a516f
commit 3aacb91
Show file tree

Hide file tree

Showing 15 changed files with 60 additions and 761 deletions.
diff --git a/__init__.py b/__init__.py
@@ -1,17 +1,13 @@
+# Thanks to RyanOnTheInside, KJNodes, MTB, Fill, Akatz, their works helped me a lot
 from .node_configs import CombinedMeta
 from collections import OrderedDict
 
-# credit to RyanOnTheInside, KJNodes, MTB, Fill, Akatz, their works helped me a lot
-
-
-#allows for central management and inheritance of class variables for help documentation
-from collections import OrderedDict
-
+# Allows for central management and inheritance of class variables for help documentation
 class Yvann(metaclass=CombinedMeta):
     @classmethod
     def get_description(cls):
-        footer = "Code : [Yvann GitHub](https://github.com/yvann-ba)\n"
-        footer += "Tutorials and Workflows: [Yvann Youtube](https://www.youtube.com/@yvann.mp4)\n\n"
+        footer = "Code: [Yvann GitHub](https://github.com/yvann-ba)\n"
+        footer += "👁️ Tutorials and Workflows: [Yvann Youtube](https://www.youtube.com/@yvann.mp4)\n\n"
 
         desc = ""
 
@@ -39,20 +35,16 @@ def get_description(cls):
 
         desc += footer
         return desc
-   
+
 from .nodes.audio.Audio_Drums_Analysis_Yvann import Audio_Drums_Analysis_Yvann
 from .nodes.audio.Audio_Vocals_Analysis_Yvann import Audio_Vocals_Analysis_Yvann
 from .nodes.audio.Audio_Analysis_Yvann import Audio_Analysis_Yvann
-from .nodes.audio.previewaudio2 import PreviewAudio2
-from .nodes.audio.imagemask import ImageAndMaskPreview2
 
 
 NODE_CLASS_MAPPINGS = {
     "Audio Drums Analysis | Yvann": Audio_Drums_Analysis_Yvann,
     "Audio Vocals Analysis | Yvann": Audio_Vocals_Analysis_Yvann,
     "Audio Analysis | Yvann": Audio_Analysis_Yvann,
-    "Preview Audio | Yvann": PreviewAudio2,
-    "Image And Mask Preview | Yvann": ImageAndMaskPreview2,
 }
 
 WEB_DIRECTORY = "./web/js"
@@ -61,8 +53,6 @@ def get_description(cls):
     "Audio Drums Analysis | Yvann": "Audio Drums Analysis | Yvann",
     "Audio Vocals Analysis | Yvann": "Audio Vocals Analysis | Yvann",
     "Audio Analysis | Yvann": "Audio Analysis | Yvann",
-    "Preview Audio | Yvann": "Preview Audio | Yvann",
-    "Image And Mask Preview | Yvann": "Image And Mask Preview | Yvann",
 }
 
 Yvann_Print = """
@@ -71,23 +61,17 @@ def get_description(cls):
 print("\033[38;5;195m" + Yvann_Print + "\033[38;5;222m" + " : Loaded\n" + "\033[0m")
 
 
-
-
 from aiohttp import web
 from server import PromptServer
 from pathlib import Path
 
 if hasattr(PromptServer, "instance"):
-
     # NOTE: we add an extra static path to avoid comfy mechanism
     # that loads every script in web. 
-
     PromptServer.instance.app.add_routes(
         [web.static("/yvann_web_async", (Path(__file__).parent.absolute() / "yvann_web_async").as_posix())]
     )
 
-
-
 for node_name, node_class in NODE_CLASS_MAPPINGS.items():
     if hasattr(node_class, 'get_description'):
         desc = node_class.get_description()

diff --git a/node_configs.py b/node_configs.py
@@ -18,33 +18,62 @@ def add_node_config(node_name, config):
 
 add_node_config("Audio_Analysis_Yvann", {
     "BASE_DESCRIPTION": """
-##Parameters
-- `video_frames`: Input video frames to be processed.
-- `audio`: Input audio to be processed.
-- `frame_rate`: Frame rate of the video.
-- `smoothing_factor`: Smoothing factor for the audio analysis
-- `global_intensity`: Global intensity for the audio analysis
+
+#### Analyzes video audio to generate audio-reactive weights and masks
+
+#### Parameters:
+- `smoothing_factor`: Smooths audio weights (higher values = smoother transitions)
+- `global_intensity`: Adjusts overall intensity of audio weights
+
+#### Inputs:
+- `video_frames`: Input video
+- `audio`: Video audio
+
+#### Outputs:
+- `Audio`: Same as input
+- `Audio Weights`: Audio-reactive float values for each frame, usable with IPAdapter, CN, etc.
+- `Audio Masks`: Audio weights converted to masks usable with any mask input: Latent Noise Mask, CN, IPadapter etc..
+- `Weights Graph`: Visual Graph of "Audio Weights", help to adjust smoothing_factor/global_intensity
 """
 })
 
 add_node_config("Audio_Drums_Analysis_Yvann", {
     "BASE_DESCRIPTION": """
-##Parameters
-- `video_frames`: Input video frames to be processed.
-- `audio`: Input audio to be processed.
-- `frame_rate`: Frame rate of the video.
-- `smoothing_factor`: Smoothing factor for the audio analysis
-- `global_intensity`: Global intensity for the audio analysis
+#### Isolates and analyzes drums sounds in video audio to generate drum-reactive weights and masks
+
+#### Parameters:
+- `smoothing_factor`: Smooths drums audio weights (higher values = smoother transitions)
+- `global_intensity`: Adjusts overall intensity of drums audio weights
+
+#### Inputs:
+- `video_frames`: Input video
+- `audio`: Video audio
+
+#### Outputs:
+- `Drums Audio`: Isolated drums audio from input
+- `Drums Weights`: Drums-reactive float values for each frame, usable with IPAdapter, CN, etc.
+- `Drums Masks`: Drums weights converted to masks usable with any mask input: Latent Noise Mask, CN, IPadapter etc..
+- `Weights Graph`: Visual Graph of "Drums Weights",help to adjust smoothing_factor/global_intensity
 """
 })
 
 add_node_config("Audio_Vocals_Analysis_Yvann", {
     "BASE_DESCRIPTION": """
-##Parameters
-- `video_frames`: Input video frames to be processed.
-- `audio`: Input audio to be processed.
-- `frame_rate`: Frame rate of the video.
-- `smoothing_factor`: Smoothing factor for the audio analysis
-- `global_intensity`: Global intensity for the audio analysis
+    
+#### Isolates and analyzes vocal sounds in video audio to generate vocal-reactive weights and masks
+
+#### Parameters:
+- `smoothing_factor`: Smooths vocals audio weights (higher values = smoother transitions)
+- `global_intensity`: Adjusts overall intensity of vocals audio weights
+
+#### Inputs:
+- `video_frames`: Input video
+- `audio`: Video audio
+
+#### Outputs:
+- `Vocals Audio`: Isolated vocals audio from input
+- `Vocals Weights`: Vocals-reactive float values for each frame, usable with IPAdapter, CN, etc.
+- `Vocals Masks`: Vocals weights converted to masks usable with any mask input: Latent Noise Mask, CN, IPadapter etc..
+- `Weights Graph`: Visual Graph of "Vocals Weights", help to adjust smoothing_factor/global_intensity
 """
 })
diff --git a/nodes/audio/Audio_Analysis_Yvann.py b/nodes/audio/Audio_Analysis_Yvann.py
@@ -4,7 +4,6 @@
 import tempfile
 import numpy as np
 from PIL import Image
-import pandas as pd
 from ... import Yvann
 import random
 
@@ -22,7 +21,6 @@ def INPUT_TYPES(cls):
             "required": {
                 "video_frames": ("IMAGE",),
                 "audio": ("AUDIO",),
-                "frame_rate": ("INT", {"default": 30, "min": 1, "max": 60, "step": 1}),
                 "smoothing_factor": ("FLOAT", {"default": 0.25, "min": 0.0, "max": 1.0, "step": 0.01}),
                 "global_intensity": ("FLOAT", {"default": 0.0, "min": -1.0, "max": 1.0, "step": 0.01}),
             }
@@ -64,8 +62,6 @@ def generate_masks(self, input_values, width, height):
         # Ensure input_values is a list
         if isinstance(input_values, (float, int)):
             input_values = [input_values]
-        elif isinstance(input_values, pd.Series):
-            input_values = input_values.tolist()
         elif isinstance(input_values, list) and all(isinstance(item, list) for item in input_values):
             input_values = [item for sublist in input_values for item in sublist]
 
@@ -79,7 +75,7 @@ def generate_masks(self, input_values, width, height):
 
         return masks_out
 
-    def process_audio(self, audio, video_frames, frame_rate, smoothing_factor, global_intensity, prompt=None, filename_prefix="ComfyUI", extra_pnginfo=None):
+    def process_audio(self, audio, video_frames, smoothing_factor, global_intensity, prompt=None, filename_prefix="ComfyUI", extra_pnginfo=None):
         waveform = audio['waveform']
         sample_rate = audio['sample_rate']
 
@@ -98,7 +94,6 @@ def process_audio(self, audio, video_frames, frame_rate, smoothing_factor, globa
         processed_audio = {
             'waveform': waveform.cpu(),
             'sample_rate': sample_rate,
-            'frame_rate': frame_rate
         }
 
         audio_weights = self._rms_energy(waveform, num_frames, samples_per_frame)

diff --git a/nodes/audio/Audio_Drums_Analysis_Yvann.py b/nodes/audio/Audio_Drums_Analysis_Yvann.py
@@ -5,9 +5,6 @@
 import tempfile
 import numpy as np
 from PIL import Image
-import librosa
-from nodes import SaveImage
-import pandas as pd
 from ... import Yvann
 
 class AudioNodeBase(Yvann):
@@ -20,7 +17,6 @@ def INPUT_TYPES(cls):
             "required": {
                 "video_frames": ("IMAGE",),
                 "audio": ("AUDIO",),
-                "frame_rate": ("INT", {"default": 30, "min": 1, "max": 60, "step": 1}),
                 "smoothing_factor": ("FLOAT", {"default": 0.25, "min": 0.0, "max": 1.0, "step": 0.01}),
                 "global_intensity": ("FLOAT", {"default": 0.0, "min": -1.0, "max": 1.0, "step": 0.01}),
             }
@@ -97,7 +93,7 @@ def generate_masks(self, input_values, width, height):
 
         return masks_out
 
-    def process_audio(self, audio, video_frames, frame_rate, smoothing_factor, global_intensity):
+    def process_audio(self, audio, video_frames, smoothing_factor, global_intensity):
         model = self.download_and_load_model()
 
         waveform = audio['waveform']
@@ -126,7 +122,6 @@ def process_audio(self, audio, video_frames, frame_rate, smoothing_factor, globa
         drums_audio = {
             'waveform': drums_waveform.cpu(),  # Move back to CPU
             'sample_rate': sample_rate,
-            'frame_rate': frame_rate
         }
 
         drums_weights = self._rms_energy(drums_waveform.squeeze(0), num_frames, samples_per_frame)

diff --git a/nodes/audio/Audio_Vocals_Analysis_Yvann.py b/nodes/audio/Audio_Vocals_Analysis_Yvann.py
@@ -5,9 +5,6 @@
 import tempfile
 import numpy as np
 from PIL import Image
-import librosa
-from nodes import SaveImage
-import pandas as pd
 from ... import Yvann
 
 class AudioNodeBase(Yvann):
@@ -21,7 +18,6 @@ def INPUT_TYPES(cls):
             "required": {
                 "video_frames": ("IMAGE",),
                 "audio": ("AUDIO",),
-                "frame_rate": ("INT", {"default": 30, "min": 1, "max": 60, "step": 1}),
                 "smoothing_factor": ("FLOAT", {"default": 0.25, "min": 0.0, "max": 1.0, "step": 0.01}),
                 "global_intensity": ("FLOAT", {"default": 0.0, "min": -1.0, "max": 1.0, "step": 0.01}),
             }
@@ -86,8 +82,6 @@ def generate_masks(self, input_values, width, height):
         # Ensure input_values is a list
         if isinstance(input_values, (float, int)):
             input_values = [input_values]
-        elif isinstance(input_values, pd.Series):
-            input_values = input_values.tolist()
         elif isinstance(input_values, list) and all(isinstance(item, list) for item in input_values):
             input_values = [item for sublist in input_values for item in sublist]
 
@@ -100,7 +94,7 @@ def generate_masks(self, input_values, width, height):
 
         return masks_out
 
-    def process_audio(self, audio, video_frames, frame_rate, smoothing_factor, global_intensity):
+    def process_audio(self, audio, video_frames, smoothing_factor, global_intensity):
         model = self.download_and_load_model()
 
         waveform = audio['waveform']
@@ -129,7 +123,6 @@ def process_audio(self, audio, video_frames, frame_rate, smoothing_factor, globa
         vocals_audio = {
             'waveform': vocals_waveform.cpu(),  # Move back to CPU
             'sample_rate': sample_rate,
-            'frame_rate': frame_rate
         }
 
         vocals_weights = self._rms_energy(vocals_waveform.squeeze(0), num_frames, samples_per_frame)