Fix CorrelationTrigger

ovgenpy/ovgenpy.py

@@ -43,10 +43,10 @@ def main(wave_dir: str, master_wave: Optional[str], fps: int):
         master_wave=master_wave,
         fps=fps,
         trigger=CorrelationTrigger.Config(
-            trigger_strength=0.1,
+            trigger_strength=10,
 
-            responsiveness=0.1,
-            falloff_width=0.5,
+            responsiveness=1,
+            falloff_width=2,
         ),
         render=RendererConfig(     # todo
             1280, 720,
@@ -107,6 +107,7 @@ class Ovgen:
             for wave in self.waves:
                 sample = round(wave.smp_s * time_seconds)
                 trigger_sample = wave.trigger.get_trigger(sample)
+                print(f'- {trigger_sample}')
                 center_smps.append(trigger_sample)
 
             print(frame)

ovgenpy/triggers.py

@@ -3,6 +3,7 @@ from typing import NamedTuple, List, Dict, Any, TYPE_CHECKING
 from dataclasses import dataclass
 
 import numpy as np
+from matplotlib import pyplot as plt
 from scipy import signal
 
 from ovgenpy.renderer import MatplotlibRenderer, RendererConfig
@@ -32,13 +33,29 @@ class TriggerConfig:
         raise NotImplementedError
 
 
-SHOW_TRIGGER = True
+SHOW_TRIGGER = False
+SHOW_TRIGGER2 = False
 
 
 def lerp(x: np.ndarray, y: np.ndarray, a: float):
     return x * (1 - a) + y * a
 
 
+class Dummy:
+    def __getattr__(self, item):
+        return self
+
+    def __call__(self, *args, **kwargs):
+        return self
+
+
+def plots(nplot):
+    for i in range(nplot):
+        if SHOW_TRIGGER2:
+            yield plt.subplot(nplot, 1, i+1)
+        else:
+            yield Dummy()
+
 class CorrelationTrigger(Trigger):
     MIN_AMPLITUDE = 0.01
 
@@ -86,19 +103,47 @@ class CorrelationTrigger(Trigger):
         data = self._wave.get_around(offset, self._buffer_nsamp)
         N = len(data)
 
+        ps = plots(4)
+        next(ps).plot(data)
+
         # Add "step function" to correlation buffer
-        prev_buffer = self._buffer.copy()
-        prev_buffer[N//2:] += trigger_strength
+        halfN = N // 2
+
+        wave_period = get_period(data)
+        window = signal.gaussian(N, std = halfN // 3)
+
+        step = np.empty(N)
+        step[:halfN] = -trigger_strength / 2
+        step[halfN:] = trigger_strength / 2
+        step *= window
+
+        prev_buffer = self._buffer + step
+        next(ps).plot(prev_buffer)
 
         # Find optimal offset (within ±N//4)
         delta = N-1
         radius = N//4
 
         # Calculate correlation
-        corr = signal.correlate(prev_buffer, data)
+        """
+        If offset < optimal, we need to `offset += positive`.
+        - The peak will appear near the right of `data`.
+        
+        Either we must slide prev_buffer to the right:
+        - correlate(data, prev_buffer)
+        - trigger = offset + peak_offset
+        
+        Or we must slide data to the left (by sliding offset to the right):
+        - correlate(prev_buffer, data)
+        - trigger = offset - peak_offset
+        """
+        corr = signal.correlate(data, prev_buffer)
         assert len(corr) == 2*N - 1
+        next(ps).plot(corr)
+
         corr = corr[delta-radius : delta+radius+1]
         delta = radius
+        next(ps).plot(corr)
 
         # argmax(corr) == delta + peak_offset == (data >> peak_offset)
         # peak_offset == argmax(corr) - delta
@@ -108,7 +153,8 @@ class CorrelationTrigger(Trigger):
         # Update correlation buffer (distinct from visible area)
         aligned = self._wave.get_around(trigger, self._buffer_nsamp)
         self._update_buffer(aligned)
-
+        if SHOW_TRIGGER2:
+            plt.show()
         return trigger
 
     def _update_buffer(self, data: np.ndarray) -> None:

tests/test_trigger.py

@@ -18,21 +18,33 @@ cfg = CorrelationTrigger.Config(
 
 
 def test_trigger():
-    wave = Wave(None, 'tests/sine440.wav')
+    # wave = Wave(None, 'tests/sine440.wav')
+    wave = Wave(None, 'tests/impulse24000.wav')
+
+    iters = 5
+    plot = False
+    x = 24000 - 500
     trigger = cfg(wave, 4000)
 
-    BIG = 0.95
-    SMALL = 0.05
-    fig, axes = plt.subplots(5, gridspec_kw=dict(
-        top=BIG, right=BIG,
-        bottom=SMALL, left=SMALL,
-    ))    # type: Figure, Axes
-    fig.tight_layout()
+    if plot:
+        BIG = 0.95
+        SMALL = 0.05
+        fig, axes = plt.subplots(iters, gridspec_kw=dict(
+            top=BIG, right=BIG,
+            bottom=SMALL, left=SMALL,
+        ))    # type: Figure, Axes
+        fig.tight_layout()
+    else:
+        axes = range(iters)
 
     for i, ax in enumerate(axes):
         if i:
-            print(trigger.get_trigger(4000))
-        ax.plot(trigger._buffer, label=str(i))
-        ax.grid()
+            offset2 = trigger.get_trigger(x)
+            print(offset2)
+            assert offset2 == 24000
+        if plot:
+            ax.plot(trigger._buffer, label=str(i))
+            ax.grid()
 
-    plt.show()
+    if plot:
+        plt.show()