How do you link the output of one plugin to the input of another plugin?

[samuelbraun04]

I'm trying to run an mp3 file through two different .dll plugins then render/write the output audio. However, the audio doesn't sound like I expect it to; since when I do the same process in FL Studio 20 (another DAW), I get a different result.

Specifically, I'm trying to bass boost the audio then put a soft clipper on it, which should result in an extremely distorted audio file. However, when I do the process through dawdreamer, the file isn't nearly as distorted as it should be.

Here's my code, does anything look off?:

import dawdreamer as daw
from os import chdir, walk 
import numpy as np
import librosa
from scipy.io import wavfile

SAMPLE_RATE = 44100
BUFFER_SIZE = 512
DURATION = 20
SOFT_CLIPPER_PLUGIN = r'C:/Users/samlb/Documents/CAR_MIX/Initial Clipper.dll'
SOFT_CLIPPER_PRESETS = r'C:/Users/samlb/Documents/CAR_MIX/PRESETS/SOFT_CLIPPER'
BASS_BOOST_PLUGIN = r'C:/Users/samlb/Documents/CAR_MIX/BarkOfDog2.dll'
BASS_BOOST_PRESETS = r'C:/Users/samlb/Documents/CAR_MIX/PRESETS/BASS_BOOST'
SONG_FOLDER = r'C:/Users/samlb/Documents/megaMe'
SONG_FILE = r'C:/Users/samlb/Documents/megaMe/ericdoa - cheap liquor (Official Audio).mp3'

def load_audio_file(file_path, duration=None):
    chdir(SONG_FOLDER)
    sig, rate = librosa.load(file_path, duration=duration, mono=False, sr=SAMPLE_RATE)
    assert(rate == SAMPLE_RATE)
    return sig

engine = daw.RenderEngine(SAMPLE_RATE, BUFFER_SIZE)

chdir(BASS_BOOST_PRESETS)
bass_boost = engine.make_plugin_processor("my_bass_boost", BASS_BOOST_PLUGIN)
assert bass_boost.get_name() == "my_bass_boost"

bass_boost.load_state(BASS_BOOST_PRESETS+'/preset1')

chdir(SOFT_CLIPPER_PRESETS)
soft_clipper = engine.make_plugin_processor("my_soft_clipper", SOFT_CLIPPER_PLUGIN)
assert soft_clipper.get_name() == "my_soft_clipper"

soft_clipper.load_state(SOFT_CLIPPER_PRESETS+'/preset1')

song = load_audio_file(SONG_FILE)

playback_processor = engine.make_playback_processor("playback", song)

graph = [
    (playback_processor, []),
    (bass_boost, ["playback"])
    (soft_clipper, ["playback"])
]

engine.load_graph(graph)
engine.render(107)
audio = engine.get_audio()

chdir(r'C:\Users\samlb\Documents\iveBeenMegaed')
wavfile.write('MEGA CHEAP LIQUOR.wav', SAMPLE_RATE, audio.transpose())

[DBraun]

Looks ok, but assuming you want to connect the output of the bass boost to the input of the soft clipper, the graph should be different.

graph = [
    (playback_processor, []),
    (bass_boost, ["playback"])
    (soft_clipper, ["my_bass_boost"])
]
# or
graph = [
    (playback_processor, []),
    (bass_boost, [playback_processor.get_name()])
    (soft_clipper, [bass_boost.get_name()])
]

Also, I see you copied assert soft_clipper.get_name() == "my_soft_clipper". This was in the README just to show get_name(), but you don't have to bother checking it in production code.

[samuelbraun04]

I think you missed a comma after the (bass_boost, ["playback"]) and (bass_boost, [playback_processor.get_name()]) lines (when I did what you wrote I got a TypeError: 'tuple' object is not callable error) , but except for that it worked. Thank you! And I'll remove the assert line lol.