def pitch_bend (size_exponent = 10, factor_exponent = 2, factor=0.5):
spec = K.Supergram(size_exponent, factor_exponent)
audio = K.Audio(spec.small_size)
size = spec.super_size
fun = K.Reals(size)
for i in range(size):
fun[i] = (0.5 + i) / size * factor
bend = K.Spline(size, size, fun)
sound1 = K.Complexes()
sound2 = K.Complexes()
energy = K.Reals()
bent = K.Reals()
audio.reading(sound1)
spec.stream_fwd(sound1, energy)
bend.stream_fwd(energy, bent)
spec.stream_bwd(bent, sound2)
audio.writing(sound2)
K.validate()
audio.run_until_input()
def pitch_bend (size_exponent=10, factor_exponent=2, factor=0.5,
width = 1600, height = 800):
spec = K.Supergram(size_exponent, factor_exponent)
audio = K.Audio(spec.small_size)
height = min(height, spec.super_size)
screen = K.Screen(width, height)
split = K.Splitter()
sound1 = K.Complexes()
sound2 = K.Complexes()
sound3 = K.Complexes()
size = spec.super_size
fun = K.Reals()
for i in range(size):
fun[i] = (0.5 + i) / size * factor
bend = K.Spline(size, size, fun)
energy = K.Reals()
bent = K.Reals()
audio.reading(sound1)
split.stream(sound1, (sound2,sound3))
spec.stream_fwd(sound2, energy)
#screen.vertical_sweeping(energy[:height] / energy.max())
bend.stream_fwd(energy, bent)
screen.vertical_sweeping(bent[:height] / bent.max())
spec.stream_bwd(bent, sound)
audio.writing(sound3)
K.validate()
audio.run_until_input()
def animate_correlogram (size_exponent = 10, factor_exponent = 2): screen = K.Screen(1024,1024) spec = K.Supergram(size_exponent, factor_exponent) audio = K.Audio(spec.small_size, writing=False) scale = spec.freq_scale(screen.height / 2) rate = audio.rate / spec.small_size loud = K.Loudness(scale.size_out, rate) cgram = K.Correlogram(scale.size_out, math.exp(-1.0 / 2.0 / rate)) assert cgram.size_out == 2 * cgram.size_in sound = K.Complexes() energy = K.Reals() pitch = K.Reals() louds = K.Reals() corr = K.Reals() audio.reading(sound) spec.stream_fwd(sound, energy) scale.stream_fwd(energy, pitch) loud.stream_fwd(pitch, louds) cgram.stream_fwd(louds, corr) screen.vertical_sweeping(corr) #...stream backward K.validate() audio.run_until_input()
def octave_lower (size = (1<<10)): audio = K.Audio(size) lower = K.OctaveLower(size) sound1 = K.Complexes() sound2 = K.Complexes() audio.reading(sound1) lower.stream(sound1, sound2) audio.writing(sound2) K.validate() audio.run_until_input()
def supergram_wire (size_exponent = 9, factor_exponent = 2): spec = K.Supergram(size_exponent, factor_exponent) audio = K.Audio(spec.small_size) sound1 = K.Complexes() sound2 = K.Complexes() energy = K.Reals() audio.reading(sound1) spec.stream_fwd(sound1, energy) spec.stream_bwd(energy, sound2) audio.writing(sound2) K.validate() audio.run_until_input()
def animate_spectrogram (exponent = 10): size = 1 << exponent length = size / 2 audio = K.Audio(size, writing=False) spec = K.Spectrogram(exponent) screen = K.Screen(size, length) sound = K.Complexes() freq = K.Reals() audio.reading(sound) spec.stream_fwd(sound, freq) screen.vertical_sweeping(K.energy_to_loudness(freq)) spec.stream_bwd(freq, sound) K.validate() audio.run_until_input()
def spectrogram_wire (num_seconds = 5): exponent = 10 size = 1 << exponent audio = K.Audio(size) spec = K.Spectrogram(exponent) wire = K.Wire() sound1 = K.Complexes() sound2 = K.Complexes() freq_in = K.Reals() freq_out = K.Reals() audio.reading(sound1) spec.stream_fwd(sound1, freq_out) wire.stream(freq_in, freq_out) spec.stream_bwd(freq_in, sound2) audio.writing(sound2) K.validate() audio.run_until_input()
def animate_freq (size_exponent = 10, factor_exponent = 2): screen = K.Screen(1024,1024) spec = K.Supergram(size_exponent, factor_exponent) audio = K.Audio(spec.small_size, writing=False) scale = spec.freq_scale(screen.height, 4000) rate = audio.rate / spec.small_size loud = K.Loudness(scale.size_out, rate) sound = K.Complexes() energy = K.Reals() pitch = K.Reals() louds = K.Reals() audio.reading(sound) spec.stream_fwd(sound, energy) scale.stream_fwd(energy, pitch) loud.stream_fwd(pitch, louds) screen.vertical_sweeping(louds) #...stream backward K.validate() audio.run_until_input()
def animate_super (size_exponent = 9,
time_exponent = 2,
freq_exponent = 2):
screen = K.Screen(1024,1024)
spec = K.Supergram(size_exponent,
time_exponent,
freq_exponent, screen.height)
audio = K.Audio(spec.small_size, writing=False)
rate = audio.rate / spec.small_size
loud = K.Loudness(spec.super_size, rate)
sound = K.Complexes()
energy = K.Reals()
louds = K.Reals()
audio.reading(sound)
spec.stream_fwd(sound, energy)
loud.stream_fwd(energy, louds)
screen.vertical_sweeping(louds)
K.validate()
audio.run_until_input()
