diff --git a/animations/spectrum.py b/animations/spectrum.py
new file mode 100644
index 0000000..d1621bd
--- /dev/null
+++ b/animations/spectrum.py
@@ -0,0 +1,170 @@
+#!/usr/bin/env python2
+
+import alsaaudio
+import numpy
+import random
+import time
+import sys
+
+
+WIDTH = 6
+HEIGHT = 8
+AUDIO_RATE = 4000 # sampling rate in Hz
+WINDOW_SIZE = 8 # average fft over WINDOW_SIZE audio frames
+
+BOX_MAP = [
+			[357,  18, 369, 186, 249, 228,  51],
+			[279,   9,  57, 159, 300, 108, 204],
+			[261,  42, 183, 201, 273, 246,  15],
+			[306, 168,  24, 138, 309, 165,  39],
+			[258, 222,  87, 363, 291, 231, 243],
+			[252, 114, 180,  75, 282, 141,  33],
+			[264, 288, 120, 135, 255,  99, 105],
+			[285, 207, 102,  45, 297, 216,  63],
+		]
+
+COLORS = [
+			[50, 255, 50], [50, 255, 120], [50, 251, 255], [50, 120, 255],
+			[50, 50, 255], [180, 50, 255], [255, 50, 137], [255, 50, 50]
+		]
+
+
+channels = {}
+
+def set_box(x,y,r,g,b):
+	if x >= 0 and y >= 0 and x < WIDTH and y < HEIGHT:
+		base_address = BOX_MAP[y][x]
+		channels[base_address] = r
+		channels[base_address + 1] = g
+		channels[base_address + 2] = b
+
+
+def output_channels():
+	for channel, value in channels.items():
+		print("%d : %d" % (channel, value))
+	print("")
+	sys.stdout.flush()
+
+
+class Background:
+	"""clear the light wall to a pseudorandomly changing/fading solid color"""
+
+	def __init__(self):
+		self._current_bg_color = [ 0, 0, 0 ]
+		self._target_bg_color = [ 128, 128, 128 ]
+		self._bg_time = 0
+
+
+	def clear(self):
+		for i in range(3):
+			if self._current_bg_color[i] < self._target_bg_color[i]:
+				self._current_bg_color[i] = self._current_bg_color[i] + 1
+			elif self._current_bg_color[i] > self._target_bg_color[i]:
+				self._current_bg_color[i] = self._current_bg_color[i] - 1
+
+			if self._current_bg_color[i] > 128:
+				self._current_bg_color[i] = 128
+			elif self._current_bg_color[i] < 0:
+				self._current_bg_color[i] = 0
+
+		self._bg_time = self._bg_time + 1
+		if self._bg_time == 64:
+			for i in range(3):
+				self._target_bg_color[i] = random.randint(0, 128)
+			self._bg_time = 0
+
+		for x in range(0, WIDTH):
+			for y in range(0, HEIGHT):
+				set_box(x, y, self._current_bg_color[0], self._current_bg_color[1], self._current_bg_color[2])
+
+
+def audio_from_raw(raw):
+	"""convert bytewise signed 16bit little endian to int list"""
+	out = []
+	high = False
+	current = 0
+	for value in raw:
+		value = ord(value[0])
+		if high:
+			sign = value & 0x80
+			current = current + 256 * (value & 0x7F)
+			if sign:
+				current =  -((~current & 0x7FFF) + 1)
+			out.append(current)
+			high = False
+		else:
+			current = value
+			high = True
+	return out
+
+
+class Fft_output:
+	"""aggregate several fft'ed samples, does postprocessing to make it look nice and outputs
+	the aggregated result"""
+
+
+	def __init__(self, width, height, windowsize):
+		self._background = Background()
+		self._width = width
+		self._height = height
+		self._windowsize = windowsize
+
+		self._count = 0
+		self._data = []
+		for _ in range(width):
+			self._data.append(0)
+
+
+	def add(self, data):
+		"""add a set of fft data to the internal store, output the result if enough data is
+			available"""
+		abss = numpy.absolute(data[1:self._width+1])
+
+		for i in range(self._width):
+			self._data[i] = self._data[i] + abss[i]
+
+		self._count = self._count + 1
+		if (self._count == self._windowsize):
+			self.output_twinkl()
+			self._count = 0
+			for i in range(self._width):
+				self._data[i] = 0
+
+
+	def output_twinkl(self):
+		"""output graph to twinkl client"""
+		# correct for disproportionately large first and second column
+		self._data[0] = self._data[0] / 2.5
+		self._data[1] = self._data[1] / 1.5
+
+		abss = numpy.absolute(self._data)
+
+		self._background.clear()
+		for col in range(self._width):
+			normalized = min(int(abss[col] / (self._height * self._windowsize * 1000)), self._height)
+			color = COLORS[normalized-1]
+			for row in range(self._height - normalized, self._height):
+				set_box(col, row, color[0], color[1], color[2])
+		output_channels()
+
+
+def init_audio(rate):
+	"""init sound input to 16bit little endian and user defined sampling rate"""
+	ain = alsaaudio.PCM(alsaaudio.PCM_CAPTURE, alsaaudio.PCM_NORMAL)
+	ain.setformat(alsaaudio.PCM_FORMAT_S16_LE)
+	ain.setrate(rate)
+
+	return ain
+
+
+def main():
+	ain = init_audio(AUDIO_RATE)
+	fft_out = Fft_output(WIDTH, HEIGHT, WINDOW_SIZE)
+
+	while True:
+		data = ain.read();
+		audio = audio_from_raw(data[1])
+		fft = numpy.fft.fft(audio)
+		fft_out.add(fft)
+
+main()