7507cc0be2
Updated with a fresh gui, customisable settings, more effects, and a lot more class.
734 lines
33 KiB
Python
734 lines
33 KiB
Python
from __future__ import print_function
|
|
from __future__ import division
|
|
import time
|
|
import sys
|
|
import numpy as np
|
|
from scipy.ndimage.filters import gaussian_filter1d
|
|
from collections import deque
|
|
from qrangeslider import QRangeSlider
|
|
from qfloatslider import QFloatSlider
|
|
import config
|
|
import microphone
|
|
import dsp
|
|
import led
|
|
if config.USE_GUI:
|
|
import pyqtgraph as pg
|
|
from PyQt5.QtCore import *
|
|
from PyQt5.QtWidgets import *
|
|
|
|
class Visualizer():
|
|
def __init__(self):
|
|
self.effects = {"Scroll":self.visualize_scroll,
|
|
"Energy":self.visualize_energy,
|
|
"Spectrum":self.visualize_spectrum,
|
|
#"Power":self.visualize_power,
|
|
"Wavelength":self.visualize_wavelength,
|
|
"Beat":self.visualize_beat,
|
|
"Wave":self.visualize_wave,}
|
|
#"Auto":self.visualize_auto}
|
|
self.colors = {"Red":(255,0,0),
|
|
"Orange":(255,40,0),
|
|
"Yellow":(255,255,0),
|
|
"Green":(0,255,0),
|
|
"Blue":(0,0,255),
|
|
"Light blue":(1,247,161),
|
|
"Purple":(80,5,252),
|
|
"Pink":(255,0,178)}
|
|
self.wavelength_color_modes = {"Spectral":"rgb",
|
|
"Dancefloor":"rpb",
|
|
"Brilliance":"ywb",
|
|
"Jungle":"ryg"}
|
|
self.current_effect = "Wavelength"
|
|
# Setup for frequency detection algorithm
|
|
self.freq_channel_history = 40
|
|
self.beat_count = 0
|
|
self.freq_channels = [deque(maxlen=self.freq_channel_history) for i in range(config.N_FFT_BINS)]
|
|
self.prev_output = np.array([[0 for i in range(config.N_PIXELS)] for i in range(3)])
|
|
self.prev_spectrum = [0 for i in range(config.N_PIXELS//2)]
|
|
self.current_freq_detects = {"beat":False,
|
|
"low":False,
|
|
"mid":False,
|
|
"high":False}
|
|
self.prev_freq_detects = {"beat":0,
|
|
"low":0,
|
|
"mid":0,
|
|
"high":0}
|
|
self.detection_ranges = {"beat":(0,1),
|
|
"low":(1,int(config.N_FFT_BINS*0.2)),
|
|
"mid":(int(config.N_FFT_BINS*0.4),int(config.N_FFT_BINS*0.6)),
|
|
"high":(int(config.N_FFT_BINS*0.7),int(config.N_FFT_BINS))}
|
|
self.min_detect_amplitude = {"beat":0.7,
|
|
"low":0.5,
|
|
"mid":0.3,
|
|
"high":0.05}
|
|
# Configurable options for effects go in here.
|
|
# Usage: self.effect_opts[effect][option]
|
|
self.effect_opts = {"Energy":{"blur": 1, # Amount of blur to apply
|
|
"scale":0.9}, # Width of effect on strip
|
|
"Wave":{"color_wave": "Red", # Colour of moving bit
|
|
"wipe_len":5, # Initial length of colour bit after beat
|
|
"wipe_speed":2}, # Number of pixels added to colour bit every frame
|
|
"Wavelength":{"roll": False, # Cycle colour overlay across strip
|
|
"color_mode": "Spectral", # Colour mode of overlay (rgb, rpb, ywb, ryg)
|
|
"mirror": False} # Reflect output down centre of strip?
|
|
}
|
|
# Configurations for dynamic ui generation. Effect options can be changed by widgets created at runtime,
|
|
# meaning that you don't need to worry about the user interface - it's all done for you.
|
|
# Each effect key points to a list. Each list contains lists giving config for each option.
|
|
# Syntax: effect:[variable, label_text, ui_element, opts]
|
|
# effect - the effect which you want to change options for. MUST have a key in self.effect_opts
|
|
# variable - the key of thing you want to be changed. MUST be in self.effect_opts[effect], otherwise it won't work.
|
|
# label - the text displayed on the ui
|
|
# ui_element - how you want the variable to be changed
|
|
# opts - options for the ui element. Must be a tuple.
|
|
# UI Elements + opts:
|
|
# slider, (min, max, interval, default) (for integer values in a given range)
|
|
# float_slider, (min, max, interval, default) (for floating point values in a given range)
|
|
# checkbox, (default) (for True/False values)
|
|
# dropdown, (dict, default) (dict example see self.colors above)
|
|
#
|
|
self.dynamic_effects_config = {"Energy":[["blur", "Blur", "float_slider", (0.1,4.0,0.1,1.0)],
|
|
["scale", "Scale", "float_slider", (0.4,1.0,0.05,0.9)]],
|
|
"Wave":[["color_wave", "Wave Color", "dropdown", self.colors],
|
|
["wipe_len", "Wave Start Length", "slider", (0,config.N_PIXELS//4,1,5)],
|
|
["wipe_speed", "Wave Speed", "slider", (1,10,1,2)]],
|
|
"Wavelength":[["roll", "Roll Colors", "checkbox", False],
|
|
["color_mode", "Color Mode", "dropdown", self.wavelength_color_modes]]
|
|
}
|
|
|
|
|
|
|
|
# Setup for "Wave" (don't change these)
|
|
self.wave_wipe_count = 0
|
|
# Setup for "Wavelength" (don't change these)
|
|
self._wavelength_set_color_mode(self.effect_opts["Wavelength"]["color_mode"])
|
|
|
|
|
|
def _wavelength_set_color_mode(self, mode):
|
|
# chunks of colour gradients
|
|
self.rgb_overlay = np.zeros((3,242))
|
|
# used to construct rgb overlay. [0-255,255...] whole length of strip
|
|
_gradient_whole = [int(i*255/(config.N_PIXELS//2)) for i in range(config.N_PIXELS//2)] +\
|
|
[255 for i in range(config.N_PIXELS//2)]
|
|
# used to construct rgb overlay. [0-255,255...] 1/2 length of strip
|
|
_gradient_half = _gradient_whole[::2]
|
|
if self.wavelength_color_modes[self.effect_opts["Wavelength"]["color_mode"]] == "rgb":
|
|
self.rgb_overlay[0, :config.N_PIXELS//2] = _gradient_half[::-1]
|
|
self.rgb_overlay[1, :] = _gradient_half + _gradient_half[::-1]
|
|
self.rgb_overlay[2, :] = np.flipud(self.rgb_overlay[0])
|
|
elif self.wavelength_color_modes[self.effect_opts["Wavelength"]["color_mode"]] == "rpb":
|
|
self.rgb_overlay[0, :] = _gradient_whole[::-1]
|
|
self.rgb_overlay[2, :] = _gradient_whole
|
|
elif self.wavelength_color_modes[self.effect_opts["Wavelength"]["color_mode"]] == "ywb":
|
|
self.rgb_overlay[0, :] = _gradient_whole[::-1]
|
|
self.rgb_overlay[1, :] = 255
|
|
self.rgb_overlay[2, :] = _gradient_whole
|
|
elif self.wavelength_color_modes[self.effect_opts["Wavelength"]["color_mode"]] == "ryg":
|
|
self.rgb_overlay[0, :] = _gradient_whole[::-1]
|
|
self.rgb_overlay[1, :] = _gradient_whole
|
|
else:
|
|
raise ValueError("Colour mode '{}' not known. Leave an issue on github if you want it added!".format(mode))
|
|
self.effect_opts["Wavelength"]["color_mode"] = mode
|
|
|
|
|
|
def get_vis(self, y):
|
|
self.update_freq_channels(y)
|
|
self.detect_freqs()
|
|
self.prev_output = np.copy(self.effects[self.current_effect](y))
|
|
return self.prev_output
|
|
|
|
def _split_equal(self, value, parts):
|
|
value = float(value)
|
|
return [int(round(i*value/parts)) for i in range(1,parts+1)]
|
|
|
|
def update_freq_channels(self, y):
|
|
for i in range(len(y)):
|
|
self.freq_channels[i].appendleft(y[i])
|
|
|
|
def detect_freqs(self):
|
|
"""
|
|
Function that updates current_freq_detects. Any visualisation algorithm can check if
|
|
there is currently a beat, low, mid, or high by querying the self.current_freq_detects dict.
|
|
"""
|
|
channel_avgs = []
|
|
differences = []
|
|
for i in range(config.N_FFT_BINS):
|
|
channel_avgs.append(sum(self.freq_channels[i])/len(self.freq_channels[i]))
|
|
differences.append(((self.freq_channels[i][0]-channel_avgs[i])*100)//channel_avgs[i])
|
|
for i in ["beat", "low", "mid", "high"]:
|
|
if any(differences[j] >= 100 and self.freq_channels[j][0] >= self.min_detect_amplitude[i]\
|
|
for j in range(*self.detection_ranges[i]))\
|
|
and (time.time() - self.prev_freq_detects[i] > 0.15)\
|
|
and len(self.freq_channels[0]) == self.freq_channel_history:
|
|
self.prev_freq_detects[i] = time.time()
|
|
self.current_freq_detects[i] = True
|
|
#print(i)
|
|
else:
|
|
self.current_freq_detects[i] = False
|
|
#if self.current_freq_detects["beat"]:
|
|
# print(time.time(),"Beat")
|
|
#pass
|
|
#print(differences[0], channel_avgs[0])
|
|
|
|
#print("{1: <{0}}{2: <{0}}{4: <{0}}{4}".format(7, self.current_freq_detects["beat"],
|
|
# self.current_freq_detects["low"],
|
|
# self.current_freq_detects["mid"],
|
|
# self.current_freq_detects["high"]))
|
|
|
|
|
|
def visualize_scroll(self, y):
|
|
"""Effect that originates in the center and scrolls outwards"""
|
|
global p
|
|
y = y**2.0
|
|
gain.update(y)
|
|
y /= gain.value
|
|
y *= 255.0
|
|
r = int(np.max(y[:len(y) // 3]))
|
|
g = int(np.max(y[len(y) // 3: 2 * len(y) // 3]))
|
|
b = int(np.max(y[2 * len(y) // 3:]))
|
|
# Scrolling effect window
|
|
p[:, 1:] = p[:, :-1]
|
|
p *= 0.98
|
|
p = gaussian_filter1d(p, sigma=0.2)
|
|
# Create new color originating at the center
|
|
p[0, 0] = r
|
|
p[1, 0] = g
|
|
p[2, 0] = b
|
|
# Update the LED strip
|
|
return np.concatenate((p[:, ::-1], p), axis=1)
|
|
|
|
|
|
def visualize_energy(self, y):
|
|
"""Effect that expands from the center with increasing sound energy"""
|
|
global p
|
|
y = np.copy(y)
|
|
gain.update(y)
|
|
y /= gain.value
|
|
scale = self.effect_opts["Energy"]["scale"]
|
|
# Scale by the width of the LED strip
|
|
y *= float((config.N_PIXELS * scale) - 1)
|
|
# Map color channels according to energy in the different freq bands
|
|
r = int(np.mean(y[:len(y) // 3]**scale))
|
|
g = int(np.mean(y[len(y) // 3: 2 * len(y) // 3]**scale))
|
|
b = int(np.mean(y[2 * len(y) // 3:]**scale))
|
|
# Assign color to different frequency regions
|
|
p[0, :r] = 255.0
|
|
p[0, r:] = 0.0
|
|
p[1, :g] = 255.0
|
|
p[1, g:] = 0.0
|
|
p[2, :b] = 255.0
|
|
p[2, b:] = 0.0
|
|
p_filt.update(p)
|
|
p = np.round(p_filt.value)
|
|
# Apply blur to smooth the edges
|
|
p[0, :] = gaussian_filter1d(p[0, :], sigma=self.effect_opts["Energy"]["blur"])
|
|
p[1, :] = gaussian_filter1d(p[1, :], sigma=self.effect_opts["Energy"]["blur"])
|
|
p[2, :] = gaussian_filter1d(p[2, :], sigma=self.effect_opts["Energy"]["blur"])
|
|
# Set the new pixel value
|
|
return np.concatenate((p[:, ::-1], p), axis=1)
|
|
|
|
def visualize_wavelength(self, y):
|
|
y = np.copy(interpolate(y, config.N_PIXELS // 2))
|
|
common_mode.update(y)
|
|
diff = y - self.prev_spectrum
|
|
self.prev_spectrum = np.copy(y)
|
|
# Color channel mappings
|
|
r = r_filt.update(y - common_mode.value)
|
|
g = np.abs(diff)
|
|
b = b_filt.update(np.copy(y))
|
|
if self.effect_opts["Wavelength"]["mirror"]:
|
|
r = r.extend(r[::-1])
|
|
r = r.extend(r[::-1])
|
|
else:
|
|
# stretch (double) r so it covers the entire spectrum
|
|
r = np.array([j for i in zip(r,r) for j in i])
|
|
b = np.array([j for i in zip(b,b) for j in i])
|
|
output = [self.rgb_overlay[0]*r,self.rgb_overlay[1]*r,self.rgb_overlay[2]*r]
|
|
self.prev_spectrum = y
|
|
if self.effect_opts["Wavelength"]["roll"]:
|
|
self.rgb_overlay = np.roll(self.rgb_overlay,1,axis=1)
|
|
output[0] = gaussian_filter1d(output[0], sigma=4.0)
|
|
output[1] = gaussian_filter1d(output[1], sigma=4.0)
|
|
output[2] = gaussian_filter1d(output[2], sigma=4.0)
|
|
return output
|
|
#return np.concatenate((p[:, ::-1], p), axis=1)
|
|
|
|
def visualize_power(self, y):
|
|
"""Effect that pulses different reqions of the strip increasing sound energy"""
|
|
global p
|
|
_p = np.copy(p)
|
|
y = np.copy(interpolate(y, config.N_PIXELS // 2))
|
|
common_mode.update(y)
|
|
diff = y - self.prev_spectrum
|
|
self.prev_spectrum = np.copy(y)
|
|
# Color channel mappings
|
|
r = r_filt.update(y - common_mode.value)
|
|
g = np.abs(diff)
|
|
b = b_filt.update(np.copy(y))
|
|
# I have no idea what any of this does but it looks cool
|
|
r = [int(i*255) for i in r[::3]]
|
|
g = [int(i*255) for i in g[::3]]
|
|
b = [int(i*255) for i in b[::3]]
|
|
_p[0, 0:len(r)] = r
|
|
_p[1, len(r):len(r)+len(g)] = g
|
|
_p[2, len(r)+len(g):config.N_PIXELS] = b[:39]
|
|
p_filt.update(_p)
|
|
# Clip it into range
|
|
_p = np.clip(p, 0, 255).astype(int)
|
|
# Apply substantial blur to smooth the edges
|
|
_p[0, :] = gaussian_filter1d(_p[0, :], sigma=3.0)
|
|
_p[1, :] = gaussian_filter1d(_p[1, :], sigma=3.0)
|
|
_p[2, :] = gaussian_filter1d(_p[2, :], sigma=3.0)
|
|
self.prev_spectrum = y
|
|
return np.concatenate((_p[:, ::-1], _p), axis=1)
|
|
|
|
def visualize_spectrum(self, y):
|
|
"""Effect that maps the Mel filterbank frequencies onto the LED strip"""
|
|
global p
|
|
#print(len(y))
|
|
#print(y)
|
|
y = np.copy(interpolate(y, config.N_PIXELS // 2))
|
|
common_mode.update(y)
|
|
diff = y - self.prev_spectrum
|
|
self.prev_spectrum = np.copy(y)
|
|
# Color channel mappings
|
|
r = r_filt.update(y - common_mode.value)
|
|
g = np.abs(diff)
|
|
b = b_filt.update(np.copy(y))
|
|
# Mirror the color channels for symmetric output
|
|
r = np.concatenate((r[::-1], r))
|
|
g = np.concatenate((g[::-1], g))
|
|
b = np.concatenate((b[::-1], b))
|
|
output = np.array([r, g,b]) * 255
|
|
self.prev_spectrum = y
|
|
return output
|
|
|
|
def visualize_auto(self,y):
|
|
"""Automatically (intelligently?) cycle through effects"""
|
|
return self.visualize_beat(y) # real intelligent
|
|
|
|
def visualize_wave(self, y):
|
|
"""Effect that flashes to the beat with scrolling coloured bits"""
|
|
if self.current_freq_detects["beat"]:
|
|
output = np.array([[255 for i in range(config.N_PIXELS)] for i in range(3)])
|
|
self.wave_wipe_count = self.effect_opts["Wave"]["wipe_len"]
|
|
else:
|
|
output = np.copy(self.prev_output)
|
|
#for i in range(len(self.prev_output)):
|
|
# output[i] = np.hsplit(self.prev_output[i],2)[0]
|
|
output = np.multiply(self.prev_output,0.7)
|
|
for i in range(self.wave_wipe_count):
|
|
output[0][i]=self.colors[self.effect_opts["Wave"]["color_wave"]][0]
|
|
output[0][-i]=self.colors[self.effect_opts["Wave"]["color_wave"]][0]
|
|
output[1][i]=self.colors[self.effect_opts["Wave"]["color_wave"]][1]
|
|
output[1][-i]=self.colors[self.effect_opts["Wave"]["color_wave"]][1]
|
|
output[2][i]=self.colors[self.effect_opts["Wave"]["color_wave"]][2]
|
|
output[2][-i]=self.colors[self.effect_opts["Wave"]["color_wave"]][2]
|
|
#output = np.concatenate([output,np.fliplr(output)], axis=1)
|
|
self.wave_wipe_count += self.effect_opts["Wave"]["wipe_speed"]
|
|
if self.wave_wipe_count > config.N_PIXELS//2:
|
|
self.wave_wipe_count = config.N_PIXELS//2
|
|
return output
|
|
|
|
def visualize_beat(self, y):
|
|
"""Effect that flashes to the beat"""
|
|
if self.current_freq_detects["beat"]:
|
|
output = np.array([[255 for i in range(config.N_PIXELS)] for i in range(3)])
|
|
else:
|
|
output = np.copy(self.prev_output)
|
|
output = np.multiply(self.prev_output,0.7)
|
|
return output
|
|
|
|
|
|
class GUI(QWidget):
|
|
def __init__(self):
|
|
super().__init__()
|
|
self.initUI()
|
|
|
|
def initUI(self):
|
|
# ==================================== Set up window and wrapping layout
|
|
self.setWindowTitle("Visualization")
|
|
wrapper = QVBoxLayout()
|
|
|
|
# ========================================== Set up FPS and error labels
|
|
labels_layout = QHBoxLayout()
|
|
self.label_error = QLabel("")
|
|
self.label_fps = QLabel("")
|
|
self.label_fps.setAlignment(Qt.AlignRight | Qt.AlignVCenter)
|
|
labels_layout.addWidget(self.label_error)
|
|
labels_layout.addStretch()
|
|
labels_layout.addWidget(self.label_fps)
|
|
|
|
# ================================================== Set up graph layout
|
|
graph_view = pg.GraphicsView()
|
|
graph_layout = pg.GraphicsLayout(border=(100,100,100))
|
|
graph_view.setCentralItem(graph_layout)
|
|
# Mel filterbank plot
|
|
fft_plot = graph_layout.addPlot(title='Filterbank Output', colspan=3)
|
|
fft_plot.setRange(yRange=[-0.1, 1.2])
|
|
fft_plot.disableAutoRange(axis=pg.ViewBox.YAxis)
|
|
x_data = np.array(range(1, config.N_FFT_BINS + 1))
|
|
self.mel_curve = pg.PlotCurveItem()
|
|
self.mel_curve.setData(x=x_data, y=x_data*0)
|
|
fft_plot.addItem(self.mel_curve)
|
|
# Visualization plot
|
|
graph_layout.nextRow()
|
|
led_plot = graph_layout.addPlot(title='Visualization Output', colspan=3)
|
|
led_plot.setRange(yRange=[-5, 260])
|
|
led_plot.disableAutoRange(axis=pg.ViewBox.YAxis)
|
|
# Pen for each of the color channel curves
|
|
r_pen = pg.mkPen((255, 30, 30, 200), width=4)
|
|
g_pen = pg.mkPen((30, 255, 30, 200), width=4)
|
|
b_pen = pg.mkPen((30, 30, 255, 200), width=4)
|
|
# Color channel curves
|
|
self.r_curve = pg.PlotCurveItem(pen=r_pen)
|
|
self.g_curve = pg.PlotCurveItem(pen=g_pen)
|
|
self.b_curve = pg.PlotCurveItem(pen=b_pen)
|
|
# Define x data
|
|
x_data = np.array(range(1, config.N_PIXELS + 1))
|
|
self.r_curve.setData(x=x_data, y=x_data*0)
|
|
self.g_curve.setData(x=x_data, y=x_data*0)
|
|
self.b_curve.setData(x=x_data, y=x_data*0)
|
|
# Add curves to plot
|
|
led_plot.addItem(self.r_curve)
|
|
led_plot.addItem(self.g_curve)
|
|
led_plot.addItem(self.b_curve)
|
|
|
|
# ================================================= Set up button layout
|
|
label_active = QLabel("Active Effect")
|
|
button_grid = QGridLayout()
|
|
buttons = {}
|
|
connecting_funcs = {}
|
|
grid_width = 4
|
|
i = 0
|
|
j = 0
|
|
# Dynamically layout buttons and connect them to the visualisation effects
|
|
def connect_generator(effect):
|
|
def func():
|
|
visualizer.current_effect = effect
|
|
func.__name__ = effect
|
|
return func
|
|
# Where the magic happens
|
|
for effect in visualizer.effects:
|
|
connecting_funcs[effect] = connect_generator(effect)
|
|
buttons[effect] = QPushButton(effect)
|
|
buttons[effect].clicked.connect(connecting_funcs[effect])
|
|
button_grid.addWidget(buttons[effect], j, i)
|
|
i += 1
|
|
if i % grid_width == 0:
|
|
i = 0
|
|
j += 1
|
|
|
|
# ============================================== Set up frequency slider
|
|
# Frequency range label
|
|
label_slider = QLabel("Frequency Range")
|
|
# Frequency slider
|
|
def freq_slider_change(tick):
|
|
minf = freq_slider.tickValue(0)**2.0 * (config.MIC_RATE / 2.0)
|
|
maxf = freq_slider.tickValue(1)**2.0 * (config.MIC_RATE / 2.0)
|
|
t = 'Frequency range: {:.0f} - {:.0f} Hz'.format(minf, maxf)
|
|
freq_label.setText(t)
|
|
config.MIN_FREQUENCY = minf
|
|
config.MAX_FREQUENCY = maxf
|
|
dsp.create_mel_bank()
|
|
def set_freq_min():
|
|
config.MIN_FREQUENCY = freq_slider.start()
|
|
dsp.create_mel_bank()
|
|
def set_freq_max():
|
|
config.MAX_FREQUENCY = freq_slider.end()
|
|
dsp.create_mel_bank()
|
|
freq_slider = QRangeSlider()
|
|
freq_slider.show()
|
|
freq_slider.setMin(0)
|
|
freq_slider.setMax(20000)
|
|
freq_slider.setRange(config.MIN_FREQUENCY, config.MAX_FREQUENCY)
|
|
freq_slider.setBackgroundStyle('background: qlineargradient(x1:0, y1:0, x2:0, y2:1, stop:0 #222, stop:1 #333);')
|
|
freq_slider.setSpanStyle('background: qlineargradient(x1:0, y1:0, x2:0, y2:1, stop:0 #282, stop:1 #393);')
|
|
freq_slider.setDrawValues(True)
|
|
freq_slider.endValueChanged.connect(set_freq_max)
|
|
freq_slider.startValueChanged.connect(set_freq_min)
|
|
freq_slider.setStyleSheet("""
|
|
QRangeSlider * {
|
|
border: 0px;
|
|
padding: 0px;
|
|
}
|
|
QRangeSlider > QSplitter::handle {
|
|
background: #fff;
|
|
}
|
|
QRangeSlider > QSplitter::handle:vertical {
|
|
height: 3px;
|
|
}
|
|
QRangeSlider > QSplitter::handle:pressed {
|
|
background: #ca5;
|
|
}
|
|
""")
|
|
|
|
# ============================================ Set up option tabs layout
|
|
label_options = QLabel("Effect Options")
|
|
opts_tabs = QTabWidget()
|
|
# Dynamically set up tabs
|
|
tabs = {}
|
|
grid_layouts = {}
|
|
self.grid_layout_widgets = {}
|
|
options = visualizer.effect_opts.keys()
|
|
for effect in visualizer.effects:
|
|
# Make the tab
|
|
self.grid_layout_widgets[effect] = {}
|
|
tabs[effect] = QWidget()
|
|
grid_layouts[effect] = QGridLayout()
|
|
tabs[effect].setLayout(grid_layouts[effect])
|
|
opts_tabs.addTab(tabs[effect],effect)
|
|
# These functions make functions for the dynamic ui generation
|
|
# YOU WANT-A DYNAMIC I GIVE-A YOU DYNAMIC!
|
|
def gen_slider_valuechanger(effect, key):
|
|
def func():
|
|
visualizer.effect_opts[effect][key] = self.grid_layout_widgets[effect][key].value()
|
|
return func
|
|
def gen_float_slider_valuechanger(effect, key):
|
|
def func():
|
|
visualizer.effect_opts[effect][key] = self.grid_layout_widgets[effect][key].slider_value
|
|
return func
|
|
def gen_combobox_valuechanger(effect, key):
|
|
def func():
|
|
visualizer.effect_opts[effect][key] = self.grid_layout_widgets[effect][key].currentText()
|
|
visualizer._wavelength_set_color_mode(visualizer.effect_opts[effect][key])
|
|
return func
|
|
def gen_checkbox_valuechanger(effect, key):
|
|
def func():
|
|
visualizer.effect_opts[effect][key] = self.grid_layout_widgets[effect][key].isChecked()
|
|
return func
|
|
# Dynamically generate ui for settings
|
|
if effect in visualizer.dynamic_effects_config:
|
|
i = 0
|
|
connecting_funcs[effect] = {}
|
|
for key, label, ui_element, opts in visualizer.dynamic_effects_config[effect][:]:
|
|
if ui_element == "slider":
|
|
connecting_funcs[effect][key] = gen_slider_valuechanger(effect, key)
|
|
self.grid_layout_widgets[effect][key] = QSlider(Qt.Horizontal)
|
|
self.grid_layout_widgets[effect][key].setMinimum(opts[0])
|
|
self.grid_layout_widgets[effect][key].setMaximum(opts[1])
|
|
self.grid_layout_widgets[effect][key].setValue(opts[2])
|
|
self.grid_layout_widgets[effect][key].valueChanged.connect(
|
|
connecting_funcs[effect][key])
|
|
grid_layouts[effect].addWidget(QLabel(label),i,0)
|
|
grid_layouts[effect].addWidget(self.grid_layout_widgets[effect][key],i,1)
|
|
elif ui_element == "float_slider":
|
|
connecting_funcs[effect][key] = gen_float_slider_valuechanger(effect, key)
|
|
self.grid_layout_widgets[effect][key] = QFloatSlider(*opts)
|
|
self.grid_layout_widgets[effect][key].valueChanged.connect(
|
|
connecting_funcs[effect][key])
|
|
grid_layouts[effect].addWidget(QLabel(label),i,0)
|
|
grid_layouts[effect].addWidget(self.grid_layout_widgets[effect][key],i,1)
|
|
elif ui_element == "dropdown":
|
|
connecting_funcs[effect][key] = gen_combobox_valuechanger(effect, key)
|
|
self.grid_layout_widgets[effect][key] = QComboBox()
|
|
self.grid_layout_widgets[effect][key].addItems(opts.keys())
|
|
self.grid_layout_widgets[effect][key].currentIndexChanged.connect(
|
|
connecting_funcs[effect][key])
|
|
grid_layouts[effect].addWidget(QLabel(label),i,0)
|
|
grid_layouts[effect].addWidget(self.grid_layout_widgets[effect][key],i,1)
|
|
elif ui_element == "checkbox":
|
|
connecting_funcs[effect][key] = gen_checkbox_valuechanger(effect, key)
|
|
self.grid_layout_widgets[effect][key] = QCheckBox()
|
|
#self.grid_layout_widgets[effect][key].addItems(opts.keys())
|
|
self.grid_layout_widgets[effect][key].stateChanged.connect(
|
|
connecting_funcs[effect][key])
|
|
grid_layouts[effect].addWidget(QLabel(label),i,0)
|
|
grid_layouts[effect].addWidget(self.grid_layout_widgets[effect][key],i,1)
|
|
i += 1
|
|
#visualizer.effect_settings[effect]
|
|
else:
|
|
grid_layouts[effect].addWidget(QLabel("No customisable options for this effect :("),0,0)
|
|
|
|
|
|
|
|
# ============================================= Add layouts into wrapper
|
|
self.setLayout(wrapper)
|
|
wrapper.addLayout(labels_layout)
|
|
wrapper.addWidget(graph_view)
|
|
wrapper.addWidget(label_active)
|
|
wrapper.addLayout(button_grid)
|
|
wrapper.addWidget(label_slider)
|
|
wrapper.addWidget(freq_slider)
|
|
wrapper.addWidget(label_options)
|
|
wrapper.addWidget(opts_tabs)
|
|
self.show()
|
|
|
|
|
|
def frames_per_second():
|
|
"""Return the estimated frames per second
|
|
|
|
Returns the current estimate for frames-per-second (FPS).
|
|
FPS is estimated by measured the amount of time that has elapsed since
|
|
this function was previously called. The FPS estimate is low-pass filtered
|
|
to reduce noise.
|
|
|
|
This function is intended to be called one time for every iteration of
|
|
the program's main loop.
|
|
|
|
Returns
|
|
-------
|
|
fps : float
|
|
Estimated frames-per-second. This value is low-pass filtered
|
|
to reduce noise.
|
|
"""
|
|
global _time_prev, _fps
|
|
time_now = time.time() * 1000.0
|
|
dt = time_now - _time_prev
|
|
_time_prev = time_now
|
|
if dt == 0.0:
|
|
return _fps.value
|
|
return _fps.update(1000.0 / dt)
|
|
|
|
def memoize(function):
|
|
"""Provides a decorator for memoizing functions"""
|
|
from functools import wraps
|
|
memo = {}
|
|
|
|
@wraps(function)
|
|
def wrapper(*args):
|
|
if args in memo:
|
|
return memo[args]
|
|
else:
|
|
rv = function(*args)
|
|
memo[args] = rv
|
|
return rv
|
|
return wrapper
|
|
|
|
@memoize
|
|
def _normalized_linspace(size):
|
|
return np.linspace(0, 1, size)
|
|
|
|
def interpolate(y, new_length):
|
|
"""Intelligently resizes the array by linearly interpolating the values
|
|
|
|
Parameters
|
|
----------
|
|
y : np.array
|
|
Array that should be resized
|
|
|
|
new_length : int
|
|
The length of the new interpolated array
|
|
|
|
Returns
|
|
-------
|
|
z : np.array
|
|
New array with length of new_length that contains the interpolated
|
|
values of y.
|
|
"""
|
|
if len(y) == new_length:
|
|
return y
|
|
x_old = _normalized_linspace(len(y))
|
|
x_new = _normalized_linspace(new_length)
|
|
z = np.interp(x_new, x_old, y)
|
|
return z
|
|
|
|
def microphone_update(audio_samples):
|
|
global y_roll, prev_rms, prev_exp, prev_fps_update
|
|
# Normalize samples between 0 and 1
|
|
y = audio_samples / 2.0**15
|
|
# Construct a rolling window of audio samples
|
|
y_roll[:-1] = y_roll[1:]
|
|
y_roll[-1, :] = np.copy(y)
|
|
y_data = np.concatenate(y_roll, axis=0).astype(np.float32)
|
|
|
|
vol = np.max(np.abs(y_data))
|
|
if vol < config.MIN_VOLUME_THRESHOLD:
|
|
if config.USE_GUI:
|
|
gui.label_error.setText("No audio input. Volume below threshold.")
|
|
else:
|
|
print("No audio input. Volume below threshold. Volume: {}".format(vol))
|
|
visualizer.prev_output = np.multiply(visualizer.prev_output,0.95)
|
|
led.pixels = visualizer.prev_output
|
|
led.update()
|
|
else:
|
|
# Transform audio input into the frequency domain
|
|
N = len(y_data)
|
|
N_zeros = 2**int(np.ceil(np.log2(N))) - N
|
|
# Pad with zeros until the next power of two
|
|
y_data *= fft_window
|
|
y_padded = np.pad(y_data, (0, N_zeros), mode='constant')
|
|
YS = np.abs(np.fft.rfft(y_padded)[:N // 2])
|
|
# Construct a Mel filterbank from the FFT data
|
|
mel = np.atleast_2d(YS).T * dsp.mel_y.T
|
|
# Scale data to values more suitable for visualization
|
|
# mel = np.sum(mel, axis=0)
|
|
mel = np.sum(mel, axis=0)
|
|
mel = mel**2.0
|
|
# Gain normalization
|
|
mel_gain.update(np.max(gaussian_filter1d(mel, sigma=1.0)))
|
|
mel /= mel_gain.value
|
|
mel = mel_smoothing.update(mel)
|
|
# Map filterbank output onto LED strip
|
|
led.pixels = visualizer.get_vis(mel)
|
|
led.update()
|
|
if config.USE_GUI:
|
|
# Plot filterbank output
|
|
x = np.linspace(config.MIN_FREQUENCY, config.MAX_FREQUENCY, len(mel))
|
|
gui.mel_curve.setData(x=x, y=fft_plot_filter.update(mel))
|
|
gui.label_error.setText("")
|
|
if config.USE_GUI:
|
|
fps = frames_per_second()
|
|
if time.time() - 0.5 > prev_fps_update:
|
|
prev_fps_update = time.time()
|
|
app.processEvents()
|
|
# Plot the color channels
|
|
gui.r_curve.setData(y=led.pixels[0])
|
|
gui.g_curve.setData(y=led.pixels[1])
|
|
gui.b_curve.setData(y=led.pixels[2])
|
|
# Update fps counter
|
|
gui.label_fps.setText('{:.0f} / {:.0f} FPS'.format(fps, config.FPS))
|
|
if config.DISPLAY_FPS:
|
|
print('FPS {:.0f} / {:.0f}'.format(fps, config.FPS))
|
|
|
|
# Initialise visualiser and GUI
|
|
visualizer = Visualizer()
|
|
if config.USE_GUI:
|
|
# Create GUI window
|
|
app = QApplication([])
|
|
app.setApplicationName('Visualization')
|
|
gui = GUI()
|
|
app.processEvents()
|
|
|
|
# Initialise filter stuff
|
|
fft_plot_filter = dsp.ExpFilter(np.tile(1e-1, config.N_FFT_BINS),
|
|
alpha_decay=0.5, alpha_rise=0.99)
|
|
mel_gain = dsp.ExpFilter(np.tile(1e-1, config.N_FFT_BINS),
|
|
alpha_decay=0.01, alpha_rise=0.99)
|
|
mel_smoothing = dsp.ExpFilter(np.tile(1e-1, config.N_FFT_BINS),
|
|
alpha_decay=0.5, alpha_rise=0.99)
|
|
volume = dsp.ExpFilter(config.MIN_VOLUME_THRESHOLD,
|
|
alpha_decay=0.02, alpha_rise=0.02)
|
|
fft_window = np.hamming(int(config.MIC_RATE / config.FPS) * config.N_ROLLING_HISTORY)
|
|
prev_fps_update = time.time()
|
|
|
|
# Initialise more filter stuff
|
|
r_filt = dsp.ExpFilter(np.tile(0.01, config.N_PIXELS // 2),
|
|
alpha_decay=0.2, alpha_rise=0.99)
|
|
g_filt = dsp.ExpFilter(np.tile(0.01, config.N_PIXELS // 2),
|
|
alpha_decay=0.05, alpha_rise=0.3)
|
|
b_filt = dsp.ExpFilter(np.tile(0.01, config.N_PIXELS // 2),
|
|
alpha_decay=0.1, alpha_rise=0.5)
|
|
common_mode = dsp.ExpFilter(np.tile(0.01, config.N_PIXELS // 2),
|
|
alpha_decay=0.99, alpha_rise=0.01)
|
|
p_filt = dsp.ExpFilter(np.tile(1, (3, config.N_PIXELS // 2)),
|
|
alpha_decay=0.1, alpha_rise=0.99)
|
|
p = np.tile(1.0, (3, config.N_PIXELS // 2))
|
|
gain = dsp.ExpFilter(np.tile(0.01, config.N_FFT_BINS),
|
|
alpha_decay=0.001, alpha_rise=0.99)
|
|
|
|
# The previous time that the frames_per_second() function was called
|
|
_time_prev = time.time() * 1000.0
|
|
# The low-pass filter used to estimate frames-per-second
|
|
_fps = dsp.ExpFilter(val=config.FPS, alpha_decay=0.2, alpha_rise=0.2)
|
|
|
|
|
|
# Number of audio samples to read every time frame
|
|
samples_per_frame = int(config.MIC_RATE / config.FPS)
|
|
# Array containing the rolling audio sample window
|
|
y_roll = np.random.rand(config.N_ROLLING_HISTORY, samples_per_frame) / 1e16
|
|
# Initialize LEDs
|
|
led.update()
|
|
# Start listening to live audio stream
|
|
microphone.start_stream(microphone_update)
|