Merge pull request #21 from FoamyGuy/master

adding wave file synth example

Author: ladyada

examples/trellism4_wavefile_synth/notes/sawtooth/A3.wav

@@ -0,0 +1,109 @@
+"""
+This synthesizer is loaded with wave files for 3 octaves of notes each in 4 different waveforms.
+It uses Mixer to play up to 7 notes at once.
+Play notes with the rainbow buttons. Change waveform types ith the white buttons in the last column.
+"""
+import board
+from audiocore import WaveFile
+from audioio import AudioOut
+from audiomixer import Mixer
+import adafruit_trellism4
+
+# trellis helper object
+trellis = adafruit_trellism4.TrellisM4Express()
+# low brightness on the neopixles
+trellis.pixels.brightness = 0.05
+# each musical note letter
+note_letters = ['C', 'D', 'E', 'F', 'G', 'A', 'B']
+# colors of the rainbow
+colors = [
+    (255, 0, 0), (255, 127, 0), (255, 255, 0),
+    (0, 255, 0), (0, 0, 255), (56, 43, 105), (139, 0, 255)]
+
+# dictionary holding note string to wave file value.
+# e.g. {... "sined4": audioio.WaveFile(open("notes/sine/d4.wav")), ...}
+notes = {}
+
+# list of all waveform types
+WAVE_TYPES = ["sine", "square", "sawtooth", "triangle"]
+
+# current waveform type. Will get changed from the last column
+current_wave_type = "sine"
+
+# load the notes dictionary
+for wave_type in WAVE_TYPES:
+    for octave in range(3, 6):  # [3,4,5]
+        for note_letter in note_letters:
+            # note with octave e.g. a4
+            cur_note = "{}{}".format(note_letter, octave)
+            # add wave file to dictionary
+            key = "{}{}".format(wave_type, cur_note)
+            notes[key] = WaveFile(
+                open("notes/{}/{}.wav".format(wave_type, cur_note), "rb"))
+
+# main audio object
+audio = AudioOut(left_channel=board.A0, right_channel=board.A1)
+# mixer to allow pylyphonic playback
+mixer = Mixer(
+    voice_count=8, sample_rate=8000,
+    channel_count=2, bits_per_sample=16,
+    samples_signed=True)
+
+audio.play(mixer)
+
+# turn on the rainbow lights
+for i, color in enumerate(colors):
+    trellis.pixels[i, 0] = color
+    trellis.pixels[i, 1] = color
+    trellis.pixels[i, 2] = color
+
+# list of keys pressed on the previous iteration
+prev_pressed = []
+
+# voice recycling variables
+available_voices = [1, 2, 3, 4, 5, 6, 7]
+# key to voice dictionary e.g. {... (1,2):4, (1,3):3,  ...}
+used_voices = {}
+
+# waveform selector in the last column
+# default to index 0 sine
+trellis.pixels[7, 0] = (255, 255, 255)
+while True:
+    cur_keys = trellis.pressed_keys
+    # if the keys are different from previous iteration
+    if cur_keys != prev_pressed:
+        # loop over currently pressed keys
+        for key in cur_keys:
+            # if it's a note key. First 7 columns.
+            if key[0] < len(note_letters):
+                # if we aren't already playing this note and we have available voice
+                if key not in used_voices.keys() and available_voices:
+                    # build not string
+                    note_for_key = "{}{}".format(note_letters[key[0]], key[1]+3)
+                    note_to_play = "{}{}".format(current_wave_type, note_for_key)
+                    # if the note exists in the notes dictionary
+                    if note_to_play in notes:
+                        # get an available voice
+                        voice_to_use = available_voices.pop()
+                        used_voices[key] = voice_to_use
+                        # play the note
+                        mixer.play(notes[note_to_play], voice=voice_to_use, loop=True)
+            else:  # last column
+                current_wave_type = WAVE_TYPES[key[1]]
+                # turn off all last column pixels
+                for y_pixel in range(0, 4):
+                    trellis.pixels[7, y_pixel] = (0, 0, 0)
+                # turn on selected
+                trellis.pixels[7, key[1]] = (255, 255, 255)
+        if mixer.playing:
+            # loop over each note that is playing
+            for key in used_voices:
+                # if the key is no longer down
+                if key not in cur_keys:
+                    # stop playing
+                    mixer.stop_voice(used_voices[key])
+                    # recycle voice
+                    available_voices.append(used_voices[key])
+                    used_voices.pop(key, None)
+    # update variable for next iteration
+    prev_pressed = cur_keys