Updated GUI

manim_recorder/multimedia.py

@@ -14,7 +14,8 @@
 from pathlib import Path
 import sox
 import numpy as np
-import noisereduce as nr
+
+# import noisereduce as nr
 import pyaudio
 from pydub.utils import mediainfo
 from mutagen.mp3 import MP3
@@ -110,6 +111,13 @@ def __setitem__(self, key, value):
     def __iter__(self):
         return iter(self.frames)
 
+    def __float__(self):
+        """Return the total recording duration in seconds."""
+        return len(self) * self.chunk / self.rate if len(self) else 0.0
+
+    def __bool__(self):
+        return True if len(self) else False
+
     def append(self, value):
         self.frames.append(value)
 
@@ -210,7 +218,7 @@ def save_recording(self) -> None:
 
     def play_playback(self) -> None:
         """Start playback of the recorded audio."""
-        if len(self):
+        if not len(self):
             return False
         self.is_playing = True
         self.stop_playback_event.clear()  # Clear the stop event
@@ -259,15 +267,9 @@ def close(self) -> None:
         """Terminate the PyAudio session."""
         self.p.terminate()
 
-    def get_recording_duration(self) -> float:
-        """Return the total recording duration in seconds."""
-        if len(self):
-            return 0.0
-        return len(self) * self.chunk / self.rate
-
     def get_recording_format_duration(self) -> str:
         """Return the recording duration formatted as HH:MM:SS."""
-        struct_time = time.gmtime(self.get_recording_duration())
+        struct_time = time.gmtime(float(self))
         return time.strftime("%H:%M:%S", struct_time)
 
     def set_filepath(self, path: str) -> None:
@@ -371,8 +373,8 @@ def trim_silence(
     return trimmed_sound
 
 
-def suppress_noise(audio_data: np.ndarray, rate) -> np.ndarray:
-    return nr.reduce_noise(y=audio_data, sr=rate)
+# def suppress_noise(audio_data: np.ndarray, rate) -> np.ndarray:
+#     return nr.reduce_noise(y=audio_data, sr=rate)
 
 
 def normalize(audio_data: np.ndarray) -> np.ndarray:

manim_recorder/recorder/gui/__gui__.py

@@ -171,7 +171,7 @@ def initUI(self):
         self.save_button = Create_Button(
             icon=self.style().standardIcon(QStyle.SP_DialogSaveButton),
             stylesheet="modern",
-            func=self.save_audio,
+            func=self._save,
             disable=True,
             toolTip="Save (<b>Ctrl + S</b>)",
         )
@@ -208,15 +208,6 @@ def initUI(self):
         # Additional Features Layout
         feature_layout = QHBoxLayout()
 
-        self.audacity_button = Create_Button(
-            "Trim Silence",
-            lambda: Run_Audacity(self.recorder_service.__str__()),
-            stylesheet="modern",
-            disable=True,
-            toolTip="Audacity Run",
-        )
-        feature_layout.addWidget(self.audacity_button)
-
         # Run Audacity Button
         self.audacity_button = Create_Button(
             "Run Audacity",
@@ -255,7 +246,7 @@ def initUI(self):
                 "s": self.__stop,
                 "p": self.__play,
                 "t": self.__pause,
-                "Ctrl+S": self.save_audio,
+                "Ctrl+S": self._save,
                 "a": self._next,
                 "Ctrl+Q": self.close,
             },
@@ -291,12 +282,12 @@ def reset_timer(self):
 
     def update_plot(self):
         """Updates the waveform plot with the latest audio data."""
-        if self.recorder_service.frames:
+        if self.recorder_service:
             # Convert the last chunk of frames to numpy array
             data = None
 
             if self.recorder_service.is_recording:
-                data = np.frombuffer(self.recorder_service.frames[-1], dtype=np.int16)
+                data = np.frombuffer(self.recorder_service[-1], dtype=np.int16)
                 # time = np.linspace(
                 #     self.recorder_service.get_recording_duration() - 1,
                 #     self.recorder_service.get_recording_duration(),
@@ -311,7 +302,7 @@ def update_plot(self):
                 < len(self.recorder_service)
             ):
                 data = np.frombuffer(
-                    self.recorder_service.frames[
+                    self.recorder_service[
                         self.recorder_service.current_playback_frame_index
                     ],
                     dtype=np.int16,
@@ -328,7 +319,7 @@ def update_progress_bar(self):
             # current_duration = (
             #     self.recorder_service.get_recording_duration()
             # )  # Total duration in seconds
-            total_frames = len(self.recorder_service.frames)  # Total frames recorded
+            total_frames = len(self.recorder_service)  # Total frames recorded
             if total_frames > 0:
                 current_frame = (
                     self.recorder_service.current_playback_frame_index
@@ -387,18 +378,21 @@ def __stop(self):
         self.play_button.setDisabled(False)
         self.rec_button.setDisabled(False)
         self.save_button.setDisabled(False)
+        self.audacity_button.setDisabled(True)
         self.accept_button.setDisabled(False)
 
     def __rec(self):
         """Starts audio recording."""
         if not self.recorder_service.is_recording:
             self.status_bar.showMessage("Status : Recording ...")
-            self.rec_button.setDisabled(True)
-            self.play_button.setDisabled(True)
-            self.save_button.setDisabled(True)
+
             self.pause_button.setDisabled(False)
+            self.play_button.setDisabled(True)
             self.stop_button.setDisabled(False)
-
+            self.rec_button.setDisabled(True)
+            self.save_button.setDisabled(True)
+            self.audacity_button.setDisabled(True)
+            self.accept_button.setDisabled(True)
             self.recorder_service.set_device_index(self.device_index)
             self.recorder_service.set_channels(self.device_index)
             self.recorder_service.start_recording()
@@ -411,7 +405,7 @@ def _next(self):
                 show_message("Confirmation", "Do you want to save the audio file?")
                 == QMessageBox.Yes
             ):
-                self.save_audio()
+                self._save()
         self.pause_button.setDisabled(True)
         self.play_button.setDisabled(True)
         self.stop_button.setDisabled(True)
@@ -424,13 +418,13 @@ def _next(self):
         if self.communicator:
             self.communicator.accept.emit(self.recorder_service.__str__())
 
-    def save_audio(self):
+    def _save(self):
         """Saves the recorded audio to a file."""
         if self.recorder_service.is_recording:
             self.recorder_service.stop_recording()
         if self.recorder_service.is_playing:
             self.recorder_service.stop_playback()
-        if self.recorder_service.frames:
+        if self.recorder_service:
             self.recorder_service.save_recording()
             self.File_Saved = True
             self.audacity_button.setDisabled(False)

manim_recorder/tracker.py

@@ -24,10 +24,12 @@ def __init__(self, scene: Scene, data: dict, cache_dir: str, voice_id: int):
         self.data = data
         self.cache_dir = cache_dir
         self.voice_id = voice_id
-        audio_file_path = Path(cache_dir) / self.data["final_audio"]
+
         self.duration = 1
-        if audio_file_path.exists():
-            self.duration = get_duration(Path(cache_dir) / self.data["final_audio"])
+        if self.data.get("final_audio"):
+            audio_file_path = Path(cache_dir) / self.data.get("final_audio")
+            if audio_file_path.exists():
+                self.duration = get_duration(audio_file_path)
         last_t = scene.renderer.time
         if last_t is None:
             last_t = 0

manim_recorder/voiceover_scene.py

@@ -24,12 +24,14 @@ class RecorderScene(Scene):
     create_subcaption: bool
     create_script: bool
     voice_id: int = -1
+    skip: bool
 
     def set_audio_service(
         self,
         audio_service: AudioService,
         create_subcaption: bool | None = True,
         cache_dir_delete: bool = False,
+        skip: bool = False,
     ) -> None:
         """Sets the Audio service to be used for the sound. This method
         should be called before adding any sound to the scene.
@@ -47,6 +49,7 @@ def set_audio_service(
         self.audio_service = audio_service
         self.current_tracker = None
         self.create_subcaption = create_subcaption
+        self.skip = skip
 
     def add_voiceover_text(
         self,
@@ -71,6 +74,20 @@ def add_voiceover_text(
         if not hasattr(self, "audio_service"):
             raise Exception("You need to call init_sound() before adding a sound.")
 
+        if self.skip:
+            return SoundTracker(
+                self,
+                {
+                    "input_data": {
+                        "id": self.voice_id,
+                        "input_text": text,
+                        "MObject": kwargs.get("mobject"),
+                    }
+                },
+                self.audio_service.cache_dir,
+                self.voice_id,
+            )
+
         dict_ = self.audio_service._wrap_generate_from_text(
             text=text, voice_id=self.voice_id, **kwargs
         )