Apply echo decay after first repeat & improve echo buffer handling

[dcooperdalrymple]

Allow independent control of the mix level of audiodelays.Echo and the decay amount by applying echo decay only when inserting echo value back into echo_buffer.

[dcooperdalrymple]

After experimenting with BlockInput on delay_ms, I quickly discovered that it causes undesirable glitching in the audio. The only solution I could think of to fix this problem is to change the echo buffer paradigm. What I've done is implemented a buffer position and rate with 8 "sub-bits". When delay_ms is updated, only the rate of iteration over the echo_buffer (with 8 additional bits of precision) is changed. During the write process, all words between the current position and the next position are written to using a basic loop.

NOTE: In the process, I did remove the separate buffer read & write positions because I didn't think they were necessary for this application.

Here is an example script to test this new feature out. You will here pitch shifting as the rate of playback over the echo buffer changes. This is normal and only affects the output when delay_ms is altered.

import board, audiobusio, audiomixer, synthio, audiodelays, time

CHANNELS = 1
SAMPLE_RATE = 44100
MAX_DELAY = 100

i2s_bclk, i2s_lclk, i2s_data = board.GP19, board.GP20, board.GP21
audio = audiobusio.I2SOut(bit_clock=i2s_bclk, word_select=i2s_lclk, data=i2s_data)
mixer = audiomixer.Mixer(voice_count=1, channel_count=CHANNELS, sample_rate=SAMPLE_RATE, buffer_size=2048)
audio.play(mixer)

synth = synthio.Synthesizer(channel_count=CHANNELS, sample_rate=SAMPLE_RATE)
synth.envelope = synthio.Envelope(attack_time=0.02, attack_level=0.75, sustain_level=0.5, release_time=0.1)

lfo = synthio.LFO(scale=MAX_DELAY/4, offset=MAX_DELAY*3/4, rate=0.75)
synth.blocks.append(lfo)

echo1 = audiodelays.Echo(max_delay_ms=MAX_DELAY, delay_ms=lfo, decay=0.5, mix=0.5, sample_rate=SAMPLE_RATE, channel_count=CHANNELS)
echo1.play(synth)
mixer.voice[0].play(echo1)

while True:
    for notenum in (60, 64, 67, 71): # C4, E4, G4, B4
        synth.press(notenum)
        time.sleep(0.1)
        synth.release(notenum)
        time.sleep(0.15)

In addition, you can actually achieve delays far beyond the max_delay_ms parameter at the expensive of audio quality. This could be useful for users who have limited ram or are intentionally looking for a "lo-fi" delay sound. In the example below, we're getting 50x the allocated delay (10ms to 500ms). I do hear some pitch shifting during this process, so there is likely room for improvement.

import board, audiobusio, audiomixer, synthio, audiodelays, time

CHANNELS = 1
SAMPLE_RATE = 44100

i2s_bclk, i2s_lclk, i2s_data = board.GP19, board.GP20, board.GP21
audio = audiobusio.I2SOut(bit_clock=i2s_bclk, word_select=i2s_lclk, data=i2s_data)
mixer = audiomixer.Mixer(voice_count=1, channel_count=CHANNELS, sample_rate=SAMPLE_RATE, buffer_size=2048)
audio.play(mixer)

synth = synthio.Synthesizer(channel_count=CHANNELS, sample_rate=SAMPLE_RATE)
synth.envelope = synthio.Envelope(attack_time=0.02, attack_level=0.75, sustain_level=0.5, release_time=0.1)

echo1 = audiodelays.Echo(max_delay_ms=10, delay_ms=500, decay=0.75, mix=0.75, sample_rate=SAMPLE_RATE, channel_count=CHANNELS)
echo1.play(synth)
mixer.voice[0].play(echo1)

synth.press(60)
time.sleep(0.1)
synth.release(60)

# Example of manual modification of `delay_ms`
time.sleep(2)
echo1.delay_ms = 400

I know this is may be against your initial plan, @gamblor21, but I think these minor changes allow this effect to be much more powerful.

[dcooperdalrymple]

As promised, here's an example of chorus using this effect. I have to admit, it's very tricky to get just right.

import board, audiobusio, audiomixer, synthio, audiodelays, time

CHANNELS = 1
SAMPLE_RATE = 44100
MAX_DELAY = 80
DEPTH = 0.25
RATE = 0.5
MIX = 0.6

i2s_bclk, i2s_lclk, i2s_data = board.GP19, board.GP20, board.GP21
audio = audiobusio.I2SOut(bit_clock=i2s_bclk, word_select=i2s_lclk, data=i2s_data)
mixer = audiomixer.Mixer(voice_count=1, channel_count=CHANNELS, sample_rate=SAMPLE_RATE, buffer_size=2048)
audio.play(mixer)

synth = synthio.Synthesizer(channel_count=CHANNELS, sample_rate=SAMPLE_RATE)
synth.envelope = synthio.Envelope(attack_time=0.02, attack_level=0.5, sustain_level=0.5, release_time=0.1)

lfo = synthio.LFO(scale=MAX_DELAY*DEPTH/2, offset=MAX_DELAY*(1-DEPTH/2), rate=RATE)
synth.blocks.append(lfo)

echo1 = audiodelays.Echo(max_delay_ms=MAX_DELAY, delay_ms=lfo, decay=0.0, mix=MIX, sample_rate=SAMPLE_RATE, channel_count=CHANNELS)
echo1.play(synth)
mixer.voice[0].play(echo1)

for notenum in (60, 64, 67, 71): # C4, E4, G4, B4
    synth.press(notenum)

while True:
    print("on")
    echo1.mix = MIX
    time.sleep(4)
    print("off")
    echo1.mix = 0
    time.sleep(4)