ThePendulum
f7c15e924c
Ports scripts/detect_claps.py (a librosa-based onset detector from an earlier session) into clap_detection.py, shared by both the CLI script and a new GUI feature: 'Detect All Claps' (whole-file scan) and 'Detect Next Clap' (from cursor, first result only), with results shown as clickable markers on a new MarkerStrip widget above the seek slider -- clicking one seeks the player to that exact timestamp. Caught a real correctness bug while testing against the out-of-sync fixture (not just the plain in-sync one): ffmpeg's default -vn audio extraction silently drops the audio stream's own start_time and renormalizes to t=0 -- which is exactly the offset this whole tool exists to fix, so losing it meant detected timestamps pointed at the wrong track's timing entirely. Fixed by probing the true start_time via ffprobe and re-inserting it as explicit silence (adelay) before analysis. Also dropped an earlier attempt at an ffmpeg -ss seek optimization for the from-cursor case, since combining it with the start_time fix meant two independent timestamp-origin corrections stacked on each other -- correct in theory, too easy to get subtly wrong in a tool where a silent wrong-timestamp bug is the worst possible failure mode. Full-file decode + filter in Python is slower but unambiguous. Verified with a smoke test that drives the real MainWindow buttons, confirms the detected marker lands near the fixture's known beep (t=5.350s, allowing for onset-detection lag), and simulates an actual QMouseEvent click on the marker to confirm click-to-seek works end-to-end.
410 lines
16 KiB
Python
410 lines
16 KiB
Python
import sys
|
|
from pathlib import Path
|
|
from typing import Optional
|
|
|
|
from PyQt6.QtCore import Qt, QTimer, QThread, pyqtSignal
|
|
from PyQt6.QtGui import QShortcut, QKeySequence, QAction
|
|
from PyQt6.QtWidgets import (
|
|
QApplication, QMainWindow, QWidget, QVBoxLayout, QHBoxLayout,
|
|
QPushButton, QLabel, QSlider, QCheckBox, QDoubleSpinBox, QFileDialog,
|
|
QMessageBox, QPlainTextEdit, QGroupBox,
|
|
)
|
|
|
|
from player_widget import PlayerWidget
|
|
from sync_logic import compute_offset, SyncOffset
|
|
from ffmpeg_export import export_synced, FfmpegExportError, ExportResult
|
|
from clap_detection import detect_claps, ClapDetectionUnavailable, ClapEvent
|
|
from marker_strip import MarkerStrip
|
|
|
|
|
|
def format_timestamp(seconds: float) -> str:
|
|
m, s = divmod(max(seconds, 0.0), 60)
|
|
return f"{int(m):02d}:{s:06.3f}"
|
|
|
|
|
|
class ExportWorker(QThread):
|
|
finished_ok = pyqtSignal(object) # ExportResult
|
|
finished_err = pyqtSignal(str, str) # message, stderr
|
|
|
|
def __init__(self, input_path, delay_ms, output_path):
|
|
super().__init__()
|
|
self.input_path = input_path
|
|
self.delay_ms = delay_ms
|
|
self.output_path = output_path
|
|
|
|
def run(self):
|
|
try:
|
|
result: ExportResult = export_synced(self.input_path, self.delay_ms, self.output_path)
|
|
self.finished_ok.emit(result)
|
|
except FfmpegExportError as e:
|
|
self.finished_err.emit(str(e), e.stderr)
|
|
|
|
|
|
class DetectWorker(QThread):
|
|
finished_ok = pyqtSignal(list) # List[ClapEvent]
|
|
finished_err = pyqtSignal(str)
|
|
|
|
def __init__(self, input_path, start_offset=0.0, max_results=None):
|
|
super().__init__()
|
|
self.input_path = input_path
|
|
self.start_offset = start_offset
|
|
self.max_results = max_results
|
|
|
|
def run(self):
|
|
try:
|
|
events = detect_claps(
|
|
self.input_path,
|
|
start_offset=self.start_offset,
|
|
max_results=self.max_results,
|
|
)
|
|
self.finished_ok.emit(events)
|
|
except (ClapDetectionUnavailable, RuntimeError) as e:
|
|
self.finished_err.emit(str(e))
|
|
|
|
|
|
class MainWindow(QMainWindow):
|
|
def __init__(self):
|
|
super().__init__()
|
|
self.setWindowTitle("A/V Sync Marker")
|
|
self.resize(960, 720)
|
|
|
|
self.input_path: Optional[str] = None
|
|
self.video_mark: Optional[float] = None
|
|
self.audio_mark: Optional[float] = None
|
|
self.offset: Optional[SyncOffset] = None
|
|
self._export_worker: Optional[ExportWorker] = None
|
|
self._detect_worker: Optional[DetectWorker] = None
|
|
self._slider_being_dragged = False
|
|
|
|
self.player = PlayerWidget()
|
|
self.player.setMinimumHeight(400)
|
|
|
|
self._build_ui()
|
|
self._build_menu()
|
|
self._wire_shortcuts()
|
|
|
|
self._position_timer = QTimer(self)
|
|
self._position_timer.setInterval(100)
|
|
self._position_timer.timeout.connect(self._update_position_ui)
|
|
self._position_timer.start()
|
|
|
|
# -- UI construction ---------------------------------------------------
|
|
|
|
def _build_ui(self):
|
|
central = QWidget()
|
|
layout = QVBoxLayout(central)
|
|
|
|
layout.addWidget(self.player, stretch=1)
|
|
|
|
# Transport row: play/pause, frame step, seek slider, position label
|
|
transport = QHBoxLayout()
|
|
self.load_button = QPushButton("Load File…")
|
|
self.load_button.clicked.connect(self._on_load_clicked)
|
|
transport.addWidget(self.load_button)
|
|
|
|
self.play_button = QPushButton("Play")
|
|
self.play_button.clicked.connect(self._toggle_play)
|
|
transport.addWidget(self.play_button)
|
|
|
|
self.prev_frame_button = QPushButton("⏮ Frame (,)")
|
|
self.prev_frame_button.clicked.connect(lambda: self._step_frame(-1))
|
|
transport.addWidget(self.prev_frame_button)
|
|
|
|
self.next_frame_button = QPushButton("Frame ⏭ (.)")
|
|
self.next_frame_button.clicked.connect(lambda: self._step_frame(1))
|
|
transport.addWidget(self.next_frame_button)
|
|
|
|
self.mute_checkbox = QCheckBox("Mute audio")
|
|
self.mute_checkbox.toggled.connect(self._on_mute_toggled)
|
|
transport.addWidget(self.mute_checkbox)
|
|
|
|
layout.addLayout(transport)
|
|
|
|
seek_row = QHBoxLayout()
|
|
self.position_label = QLabel("00:00.000")
|
|
seek_row.addWidget(self.position_label)
|
|
|
|
slider_stack = QVBoxLayout()
|
|
slider_stack.setSpacing(0)
|
|
self.marker_strip = MarkerStrip()
|
|
self.marker_strip.markerClicked.connect(self._on_marker_clicked)
|
|
slider_stack.addWidget(self.marker_strip)
|
|
|
|
self.seek_slider = QSlider(Qt.Orientation.Horizontal)
|
|
self.seek_slider.setRange(0, 0)
|
|
self.seek_slider.sliderPressed.connect(lambda: setattr(self, "_slider_being_dragged", True))
|
|
self.seek_slider.sliderReleased.connect(self._on_slider_released)
|
|
slider_stack.addWidget(self.seek_slider)
|
|
seek_row.addLayout(slider_stack, stretch=1)
|
|
|
|
self.duration_label = QLabel("00:00.000")
|
|
seek_row.addWidget(self.duration_label)
|
|
layout.addLayout(seek_row)
|
|
|
|
# Clap/onset auto-detect row
|
|
detect_group = QGroupBox("Clap detection")
|
|
detect_layout = QHBoxLayout(detect_group)
|
|
|
|
self.detect_all_button = QPushButton("Detect All Claps (whole file)")
|
|
self.detect_all_button.setEnabled(False)
|
|
self.detect_all_button.clicked.connect(self._on_detect_all_clicked)
|
|
detect_layout.addWidget(self.detect_all_button)
|
|
|
|
self.detect_next_button = QPushButton("Detect Next Clap (from cursor)")
|
|
self.detect_next_button.setEnabled(False)
|
|
self.detect_next_button.clicked.connect(self._on_detect_next_clicked)
|
|
detect_layout.addWidget(self.detect_next_button)
|
|
|
|
self.detect_status_label = QLabel("")
|
|
detect_layout.addWidget(self.detect_status_label, stretch=1)
|
|
|
|
layout.addWidget(detect_group)
|
|
|
|
# Marker row
|
|
marker_group = QGroupBox("Sync markers")
|
|
marker_layout = QHBoxLayout(marker_group)
|
|
|
|
self.mark_video_button = QPushButton("Mark Video Event")
|
|
self.mark_video_button.clicked.connect(self._mark_video)
|
|
marker_layout.addWidget(self.mark_video_button)
|
|
self.video_mark_label = QLabel("video: —")
|
|
marker_layout.addWidget(self.video_mark_label)
|
|
|
|
self.mark_audio_button = QPushButton("Mark Audio Event")
|
|
self.mark_audio_button.clicked.connect(self._mark_audio)
|
|
marker_layout.addWidget(self.mark_audio_button)
|
|
self.audio_mark_label = QLabel("audio: —")
|
|
marker_layout.addWidget(self.audio_mark_label)
|
|
|
|
layout.addWidget(marker_group)
|
|
|
|
# Offset + export row
|
|
export_group = QGroupBox("Offset and export")
|
|
export_layout = QHBoxLayout(export_group)
|
|
|
|
self.offset_label = QLabel("Offset: — (mark both events)")
|
|
export_layout.addWidget(self.offset_label, stretch=1)
|
|
|
|
export_layout.addWidget(QLabel("Override (ms):"))
|
|
self.override_spinbox = QDoubleSpinBox()
|
|
self.override_spinbox.setRange(-600000, 600000)
|
|
self.override_spinbox.setDecimals(0)
|
|
self.override_spinbox.setEnabled(False)
|
|
export_layout.addWidget(self.override_spinbox)
|
|
|
|
self.export_button = QPushButton("Export…")
|
|
self.export_button.setEnabled(False)
|
|
self.export_button.clicked.connect(self._on_export_clicked)
|
|
export_layout.addWidget(self.export_button)
|
|
|
|
layout.addWidget(export_group)
|
|
|
|
self.log_view = QPlainTextEdit()
|
|
self.log_view.setReadOnly(True)
|
|
self.log_view.setMaximumHeight(120)
|
|
self.log_view.setPlaceholderText("ffmpeg output / warnings will appear here after export.")
|
|
layout.addWidget(self.log_view)
|
|
|
|
self.setCentralWidget(central)
|
|
|
|
def _build_menu(self):
|
|
view_menu = self.menuBar().addMenu("&View")
|
|
|
|
self.toggle_log_action = QAction("Show ffmpeg Log", self, checkable=True, checked=True)
|
|
self.toggle_log_action.toggled.connect(self.log_view.setVisible)
|
|
view_menu.addAction(self.toggle_log_action)
|
|
|
|
def _wire_shortcuts(self):
|
|
QShortcut(QKeySequence(","), self, activated=lambda: self._step_frame(-1))
|
|
QShortcut(QKeySequence("."), self, activated=lambda: self._step_frame(1))
|
|
QShortcut(QKeySequence(Qt.Key.Key_Space), self, activated=self._toggle_play)
|
|
|
|
# -- transport -----------------------------------------------------
|
|
|
|
def _on_load_clicked(self):
|
|
path, _ = QFileDialog.getOpenFileName(self, "Load video file", "", "Video files (*.mp4 *.mov *.mkv *.avi *.webm);;All files (*)")
|
|
if path:
|
|
self._load(path)
|
|
|
|
def _load(self, path: str):
|
|
self.input_path = path
|
|
self.player.load(path)
|
|
self.video_mark = None
|
|
self.audio_mark = None
|
|
self.offset = None
|
|
self.video_mark_label.setText("video: —")
|
|
self.audio_mark_label.setText("audio: —")
|
|
self.offset_label.setText("Offset: — (mark both events)")
|
|
self.export_button.setEnabled(False)
|
|
self.override_spinbox.setEnabled(False)
|
|
self.log_view.clear()
|
|
|
|
duration = self.player.mpv.duration or 0.0
|
|
self.seek_slider.setRange(0, int(duration * 1000))
|
|
self.duration_label.setText(format_timestamp(duration))
|
|
self.setWindowTitle(f"A/V Sync Marker — {Path(path).name}")
|
|
|
|
self.marker_strip.set_duration(duration)
|
|
self.marker_strip.clear()
|
|
self.detect_status_label.setText("")
|
|
self.detect_all_button.setEnabled(True)
|
|
self.detect_next_button.setEnabled(True)
|
|
|
|
def _toggle_play(self):
|
|
if self.player.mpv.pause:
|
|
self.player.play()
|
|
self.play_button.setText("Pause")
|
|
else:
|
|
self.player.pause()
|
|
self.play_button.setText("Play")
|
|
|
|
def _step_frame(self, n: int):
|
|
self.player.pause()
|
|
self.play_button.setText("Play")
|
|
self.player.step_frame(n)
|
|
self._update_position_ui()
|
|
|
|
def _on_mute_toggled(self, checked: bool):
|
|
self.player.mpv.mute = checked
|
|
|
|
def _on_slider_released(self):
|
|
target_sec = self.seek_slider.value() / 1000.0
|
|
self.player.seek(target_sec)
|
|
self._slider_being_dragged = False
|
|
self._update_position_ui()
|
|
|
|
def _update_position_ui(self):
|
|
if self.input_path is None:
|
|
return
|
|
pos = self.player.get_position()
|
|
self.position_label.setText(format_timestamp(pos))
|
|
if not self._slider_being_dragged:
|
|
self.seek_slider.blockSignals(True)
|
|
self.seek_slider.setValue(int(pos * 1000))
|
|
self.seek_slider.blockSignals(False)
|
|
|
|
# -- markers / offset -----------------------------------------------
|
|
|
|
def _mark_video(self):
|
|
self.video_mark = self.player.get_position()
|
|
self.video_mark_label.setText(f"video: {format_timestamp(self.video_mark)}")
|
|
self._recompute_offset()
|
|
|
|
def _mark_audio(self):
|
|
self.audio_mark = self.player.get_position()
|
|
self.audio_mark_label.setText(f"audio: {format_timestamp(self.audio_mark)}")
|
|
self._recompute_offset()
|
|
|
|
def _recompute_offset(self):
|
|
if self.video_mark is None or self.audio_mark is None:
|
|
return
|
|
self.offset = compute_offset(self.video_mark, self.audio_mark)
|
|
self.offset_label.setText(self.offset.describe())
|
|
self.override_spinbox.setValue(self.offset.delay_ms)
|
|
self.override_spinbox.setEnabled(True)
|
|
self.export_button.setEnabled(True)
|
|
|
|
# -- clap detection -----------------------------------------------------
|
|
|
|
def _on_marker_clicked(self, time_sec: float):
|
|
self.player.seek(time_sec)
|
|
self._update_position_ui()
|
|
|
|
def _on_detect_all_clicked(self):
|
|
self._start_detection(start_offset=0.0, max_results=None, label="Scanning whole file for claps…")
|
|
|
|
def _on_detect_next_clicked(self):
|
|
cursor = self.player.get_position()
|
|
self._start_detection(start_offset=cursor, max_results=1, label=f"Scanning from {format_timestamp(cursor)} for next clap…")
|
|
|
|
def _start_detection(self, start_offset: float, max_results, label: str):
|
|
if self.input_path is None:
|
|
return
|
|
self.detect_all_button.setEnabled(False)
|
|
self.detect_next_button.setEnabled(False)
|
|
self.detect_status_label.setText(label)
|
|
|
|
self._detect_worker = DetectWorker(self.input_path, start_offset=start_offset, max_results=max_results)
|
|
self._detect_worker.finished_ok.connect(self._on_detect_done)
|
|
self._detect_worker.finished_err.connect(self._on_detect_failed)
|
|
self._detect_worker.start()
|
|
|
|
def _on_detect_done(self, events: list):
|
|
self.detect_all_button.setEnabled(True)
|
|
self.detect_next_button.setEnabled(True)
|
|
|
|
existing = set(self.marker_strip.markers())
|
|
existing.update(e.time_sec for e in events)
|
|
self.marker_strip.set_markers(list(existing))
|
|
|
|
if not events:
|
|
self.detect_status_label.setText("No claps found.")
|
|
self.log_view.appendPlainText("Clap detection: no onsets found.")
|
|
return
|
|
|
|
self.detect_status_label.setText(f"Found {len(events)} clap(s) — click a marker above the timeline to jump to it.")
|
|
for e in events:
|
|
self.log_view.appendPlainText(f"Clap detected at {format_timestamp(e.time_sec)} (strength {e.strength:.2f})")
|
|
|
|
def _on_detect_failed(self, message: str):
|
|
self.detect_all_button.setEnabled(True)
|
|
self.detect_next_button.setEnabled(True)
|
|
self.detect_status_label.setText("Detection failed (see log).")
|
|
self.log_view.appendPlainText(f"Clap detection FAILED: {message}")
|
|
|
|
# -- export -----------------------------------------------------
|
|
|
|
def _on_export_clicked(self):
|
|
if self.input_path is None or self.offset is None:
|
|
return
|
|
delay_ms = self.override_spinbox.value()
|
|
|
|
input_p = Path(self.input_path)
|
|
suggested = str(input_p.with_name(f"{input_p.stem}_synced{input_p.suffix}"))
|
|
output_path, _ = QFileDialog.getSaveFileName(self, "Export synced file", suggested)
|
|
if not output_path:
|
|
return
|
|
|
|
self.export_button.setEnabled(False)
|
|
self.export_button.setText("Exporting…")
|
|
self.log_view.appendPlainText(f"Exporting with delay_ms={delay_ms:.0f} …")
|
|
|
|
self._export_worker = ExportWorker(self.input_path, delay_ms, output_path)
|
|
self._export_worker.finished_ok.connect(self._on_export_done)
|
|
self._export_worker.finished_err.connect(self._on_export_failed)
|
|
self._export_worker.start()
|
|
|
|
def _on_export_done(self, result: ExportResult):
|
|
self.export_button.setEnabled(True)
|
|
self.export_button.setText("Export…")
|
|
self.log_view.appendPlainText("Export finished.")
|
|
if result.stderr.strip():
|
|
# ffmpeg logs things like "sps_id out of range" on stream-copy
|
|
# of screen-recording-style sources; harmless, just surfaced
|
|
# for visibility rather than treated as failure.
|
|
self.log_view.appendPlainText("ffmpeg log:\n" + result.stderr)
|
|
QMessageBox.information(self, "Export complete", "Export finished successfully.")
|
|
|
|
def _on_export_failed(self, message: str, stderr: str):
|
|
self.export_button.setEnabled(True)
|
|
self.export_button.setText("Export…")
|
|
self.log_view.appendPlainText(f"Export FAILED: {message}\n{stderr}")
|
|
QMessageBox.critical(self, "Export failed", f"{message}\n\n{stderr[-2000:]}")
|
|
|
|
def closeEvent(self, event):
|
|
self.player.shutdown()
|
|
super().closeEvent(event)
|
|
|
|
|
|
def main():
|
|
app = QApplication(sys.argv)
|
|
window = MainWindow()
|
|
window.show()
|
|
if len(sys.argv) > 1:
|
|
window._load(sys.argv[1])
|
|
sys.exit(app.exec())
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|