RealTimeSTT 0.2.0__tar.gz → 0.2.2__tar.gz

{RealtimeSTT-0.2.0 → RealtimeSTT-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: RealtimeSTT
-Version: 0.2.0
+Version: 0.2.2
 Summary: A fast Voice Activity Detection and Transcription System
 Home-page: https://github.com/KoljaB/RealTimeSTT
 Author: Kolja Beigel
@@ -12,16 +12,6 @@ Classifier: Operating System :: OS Independent
 Requires-Python: >=3.6
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: PyAudio==0.2.14
-Requires-Dist: faster-whisper==1.0.2
-Requires-Dist: pvporcupine==1.9.5
-Requires-Dist: webrtcvad==2.0.10
-Requires-Dist: halo==0.0.31
-Requires-Dist: torch==2.3.1
-Requires-Dist: torchaudio==2.3.1
-Requires-Dist: scipy==1.12.0
-Requires-Dist: websockets==v12.0
-Requires-Dist: openwakeword==0.6.0
 
 
 # RealtimeSTT
@@ -47,7 +37,7 @@ https://github.com/KoljaB/RealtimeSTT/assets/7604638/207cb9a2-4482-48e7-9d2b-072
 
 ### Updates
 
-Latest Version: v0.2.0
+Latest Version: v0.2.2
 
 See [release history](https://github.com/KoljaB/RealtimeSTT/releases).
 
@@ -89,24 +79,9 @@ Although it is possible to run RealtimeSTT with a CPU installation only (use a s
 ### GPU Support with CUDA (recommended)
 
 
-**Install RealtimeSTT with CUDA support**  
+### Updating PyTorch for CUDA Support
 
-*For Cuda 11.8:*  
-```bash
-pip install RealtimeSTT --index-url https://download.pytorch.org/whl/cu118
-```  
-
-
-*For Cuda 12.X:*  
-```bash
-pip install RealtimeSTT --index-url https://download.pytorch.org/whl/cu121
-```
-
-This command reinstalls PyTorch and Torchaudio with explicit support for CUDA.
-
-### Updating PyTorch Only for CUDA Support
-
-If you already have RealtimeSTT installed but want to upgrade your PyTorch installation to leverage GPU support with CUDA, follow these instructions based on your specific CUDA version. This process is useful if you initially installed RealtimeSTT with only CPU support and now wish to enhance its performance with CUDA capabilities.
+To upgrade your PyTorch installation to enable GPU support with CUDA, follow these instructions based on your specific CUDA version. This is useful if you wish to enhance the performance of RealtimeSTT with CUDA capabilities.
 
 #### For CUDA 11.8:
 To update PyTorch and Torchaudio to support CUDA 11.8, use the following commands:
@@ -122,7 +97,7 @@ To update PyTorch and Torchaudio to support CUDA 12.X, execute the following:
 pip install torch==2.3.1+cu121 torchaudio==2.3.1 --index-url https://download.pytorch.org/whl/cu121
 ```
 
-Replace `2.3.1` with the version of CUDA that matches your system and requirements.
+Replace `2.3.1` with the version of PyTorch that matches your system and requirements.
 
 ### Steps That Might Be Necessary Before
 
@@ -373,6 +348,8 @@ When you initialize the `AudioToTextRecorder` class, you have various options to
 
 - **silero_use_onnx** (bool, default=False): Enables usage of the pre-trained model from Silero in the ONNX (Open Neural Network Exchange) format instead of the PyTorch format. Default is False. Recommended for faster performance.
 
+- **silero_deactivity_detection** (bool, default=False): Enables the Silero model for end-of-speech detection. More robust against background noise. Utilizes additional GPU resources but improves accuracy in noisy environments. When False, uses the default WebRTC VAD, which is more sensitive but may continue recording longer due to background sounds.
+
 - **webrtc_sensitivity** (int, default=3): Sensitivity for the WebRTC Voice Activity Detection engine ranging from 0 (least aggressive / most sensitive) to 3 (most aggressive, least sensitive). Default is 3.
 
 - **post_speech_silence_duration** (float, default=0.2): Duration in seconds of silence that must follow speech before the recording is considered to be completed. This ensures that any brief pauses during speech don't prematurely end the recording.

{RealtimeSTT-0.2.0 → RealtimeSTT-0.2.2}/README.md

@@ -22,7 +22,7 @@ https://github.com/KoljaB/RealtimeSTT/assets/7604638/207cb9a2-4482-48e7-9d2b-072
 
 ### Updates
 
-Latest Version: v0.2.0
+Latest Version: v0.2.2
 
 See [release history](https://github.com/KoljaB/RealtimeSTT/releases).
 
@@ -64,24 +64,9 @@ Although it is possible to run RealtimeSTT with a CPU installation only (use a s
 ### GPU Support with CUDA (recommended)
 
 
-**Install RealtimeSTT with CUDA support**  
+### Updating PyTorch for CUDA Support
 
-*For Cuda 11.8:*  
-```bash
-pip install RealtimeSTT --index-url https://download.pytorch.org/whl/cu118
-```  
-
-
-*For Cuda 12.X:*  
-```bash
-pip install RealtimeSTT --index-url https://download.pytorch.org/whl/cu121
-```
-
-This command reinstalls PyTorch and Torchaudio with explicit support for CUDA.
-
-### Updating PyTorch Only for CUDA Support
-
-If you already have RealtimeSTT installed but want to upgrade your PyTorch installation to leverage GPU support with CUDA, follow these instructions based on your specific CUDA version. This process is useful if you initially installed RealtimeSTT with only CPU support and now wish to enhance its performance with CUDA capabilities.
+To upgrade your PyTorch installation to enable GPU support with CUDA, follow these instructions based on your specific CUDA version. This is useful if you wish to enhance the performance of RealtimeSTT with CUDA capabilities.
 
 #### For CUDA 11.8:
 To update PyTorch and Torchaudio to support CUDA 11.8, use the following commands:
@@ -97,7 +82,7 @@ To update PyTorch and Torchaudio to support CUDA 12.X, execute the following:
 pip install torch==2.3.1+cu121 torchaudio==2.3.1 --index-url https://download.pytorch.org/whl/cu121
 ```
 
-Replace `2.3.1` with the version of CUDA that matches your system and requirements.
+Replace `2.3.1` with the version of PyTorch that matches your system and requirements.
 
 ### Steps That Might Be Necessary Before
 
@@ -348,6 +333,8 @@ When you initialize the `AudioToTextRecorder` class, you have various options to
 
 - **silero_use_onnx** (bool, default=False): Enables usage of the pre-trained model from Silero in the ONNX (Open Neural Network Exchange) format instead of the PyTorch format. Default is False. Recommended for faster performance.
 
+- **silero_deactivity_detection** (bool, default=False): Enables the Silero model for end-of-speech detection. More robust against background noise. Utilizes additional GPU resources but improves accuracy in noisy environments. When False, uses the default WebRTC VAD, which is more sensitive but may continue recording longer due to background sounds.
+
 - **webrtc_sensitivity** (int, default=3): Sensitivity for the WebRTC Voice Activity Detection engine ranging from 0 (least aggressive / most sensitive) to 3 (most aggressive, least sensitive). Default is 3.
 
 - **post_speech_silence_duration** (float, default=0.2): Duration in seconds of silence that must follow speech before the recording is considered to be completed. This ensures that any brief pauses during speech don't prematurely end the recording.

{RealtimeSTT-0.2.0 → RealtimeSTT-0.2.2}/RealtimeSTT/audio_recorder.py

@@ -115,6 +115,7 @@ class AudioToTextRecorder:
                  # Voice activation parameters
                  silero_sensitivity: float = INIT_SILERO_SENSITIVITY,
                  silero_use_onnx: bool = False,
+                 silero_deactivity_detection: bool = False,
                  webrtc_sensitivity: int = INIT_WEBRTC_SENSITIVITY,
                  post_speech_silence_duration: float = (
                      INIT_POST_SPEECH_SILENCE_DURATION
@@ -228,6 +229,12 @@ class AudioToTextRecorder:
             pre-trained model from Silero in the ONNX (Open Neural Network
             Exchange) format instead of the PyTorch format. This is
             recommended for faster performance.
+		- silero_deactivity_detection (bool, default=False): Enables the Silero
+            model for end-of-speech detection. More robust against background
+            noise. Utilizes additional GPU resources but improves accuracy in
+            noisy environments. When False, uses the default WebRTC VAD,
+            which is more sensitive but may continue recording longer due
+            to background sounds.
         - webrtc_sensitivity (int, default=WEBRTC_SENSITIVITY): Sensitivity
             for the WebRTC Voice Activity Detection engine ranging from 0
             (least aggressive / most sensitive) to 3 (most aggressive,
@@ -381,6 +388,7 @@ class AudioToTextRecorder:
         self.silero_working = False
         self.speech_end_silence_start = 0
         self.silero_sensitivity = silero_sensitivity
+        self.silero_deactivity_detection = silero_deactivity_detection
         self.listen_start = 0
         self.spinner = spinner
         self.halo = None
@@ -443,7 +451,7 @@ class AudioToTextRecorder:
         # Set device for model
         self.device = "cuda" if self.device == "cuda" and torch.cuda.is_available() else "cpu"
 
-        self.transcript_process = mp.Process(
+        self.transcript_process = self._start_thread(
             target=AudioToTextRecorder._transcription_worker,
             args=(
                 child_transcription_pipe,
@@ -459,7 +467,6 @@ class AudioToTextRecorder:
                 self.suppress_tokens
             )
         )
-        self.transcript_process.start()
 
         # Start audio data reading process
         if self.use_microphone.value:
@@ -468,7 +475,7 @@ class AudioToTextRecorder:
                          f" sample rate: {self.sample_rate}"
                          f" buffer size: {self.buffer_size}"
                          )
-            self.reader_process = mp.Process(
+            self.reader_process = self._start_thread(
                 target=AudioToTextRecorder._audio_data_worker,
                 args=(
                     self.audio_queue,
@@ -480,7 +487,6 @@ class AudioToTextRecorder:
                     self.use_microphone
                 )
             )
-            self.reader_process.start()
 
         # Initialize the realtime transcription model
         if self.enable_realtime_transcription:
@@ -517,69 +523,71 @@ class AudioToTextRecorder:
                 for _ in range(len(self.wake_words_list))
             ]
 
-            match self.wakeword_backend:
-                case 'pvp' | 'pvporcupine':
-                    try:
-                        self.porcupine = pvporcupine.create(
-                            keywords=self.wake_words_list,
-                            sensitivities=self.wake_words_sensitivities
-                        )
-                        self.buffer_size = self.porcupine.frame_length
-                        self.sample_rate = self.porcupine.sample_rate
+            if self.wakeword_backend in {'pvp', 'pvporcupine'}:
 
-                    except Exception as e:
-                        logging.exception(
-                            "Error initializing porcupine "
-                            f"wake word detection engine: {e}"
-                        )
-                        raise
+                try:
+                    self.porcupine = pvporcupine.create(
+                        keywords=self.wake_words_list,
+                        sensitivities=self.wake_words_sensitivities
+                    )
+                    self.buffer_size = self.porcupine.frame_length
+                    self.sample_rate = self.porcupine.sample_rate
 
-                    logging.debug(
-                        "Porcupine wake word detection engine initialized successfully"
+                except Exception as e:
+                    logging.exception(
+                        "Error initializing porcupine "
+                        f"wake word detection engine: {e}"
                     )
+                    raise
 
-                case 'oww' | 'openwakeword' | 'openwakewords':
-                    openwakeword.utils.download_models()
+                logging.debug(
+                    "Porcupine wake word detection engine initialized successfully"
+                )
 
-                    try:
-                        if openwakeword_model_paths:
-                            model_paths = openwakeword_model_paths.split(',')
-                            self.owwModel = Model(
-                                wakeword_models=model_paths,
-                                inference_framework=openwakeword_inference_framework
-                            )
-                            logging.info(
-                                "Successfully loaded wakeword model(s): "
-                                f"{openwakeword_model_paths}"
-                            )
-                        else:
-                            self.owwModel = Model(
-                                inference_framework=openwakeword_inference_framework)
-                        
-                        self.oww_n_models = len(self.owwModel.models.keys())
-                        if not self.oww_n_models:
-                            logging.error(
-                                "No wake word models loaded."
-                            )
-   
-                        for model_key in self.owwModel.models.keys():
-                            logging.info(
-                                "Successfully loaded openwakeword model: "
-                                f"{model_key}"
-                            )
+            elif self.wakeword_backend in {'oww', 'openwakeword', 'openwakewords'}:
+                    
+                openwakeword.utils.download_models()
 
-                    except Exception as e:
-                        logging.exception(
-                            "Error initializing openwakeword "
-                            f"wake word detection engine: {e}"
+                try:
+                    if openwakeword_model_paths:
+                        model_paths = openwakeword_model_paths.split(',')
+                        self.owwModel = Model(
+                            wakeword_models=model_paths,
+                            inference_framework=openwakeword_inference_framework
+                        )
+                        logging.info(
+                            "Successfully loaded wakeword model(s): "
+                            f"{openwakeword_model_paths}"
+                        )
+                    else:
+                        self.owwModel = Model(
+                            inference_framework=openwakeword_inference_framework)
+                    
+                    self.oww_n_models = len(self.owwModel.models.keys())
+                    if not self.oww_n_models:
+                        logging.error(
+                            "No wake word models loaded."
+                        )
+
+                    for model_key in self.owwModel.models.keys():
+                        logging.info(
+                            "Successfully loaded openwakeword model: "
+                            f"{model_key}"
                         )
-                        raise
 
-                    logging.debug(
-                        "Open wake word detection engine initialized successfully"
+                except Exception as e:
+                    logging.exception(
+                        "Error initializing openwakeword "
+                        f"wake word detection engine: {e}"
                     )
-                case _:
-                    logging.exception("Wakeword engine {} unknown/unsupported. Please specify one of: pvporcupine, openwakeword.")
+                    raise
+
+                logging.debug(
+                    "Open wake word detection engine initialized successfully"
+                )
+            
+            else:
+                logging.exception(f"Wakeword engine {self.wakeword_backend} unknown/unsupported. Please specify one of: pvporcupine, openwakeword.")
 
 
         # Setup voice activity detection model WebRTC
@@ -640,13 +648,36 @@ class AudioToTextRecorder:
         self.realtime_thread = threading.Thread(target=self._realtime_worker)
         self.realtime_thread.daemon = True
         self.realtime_thread.start()
-
+                   
         # Wait for transcription models to start
         logging.debug('Waiting for main transcription model to start')
         self.main_transcription_ready_event.wait()
         logging.debug('Main transcription model ready')
 
         logging.debug('RealtimeSTT initialization completed successfully')
+                   
+    def _start_thread(self, target=None, args=()):
+        """
+        Implement a consistent threading model across the library.
+
+        This method is used to start any thread in this library. It uses the
+        standard threading. Thread for Linux and for all others uses the pytorch
+        MultiProcessing library 'Process'.
+        Args:
+            target (callable object): is the callable object to be invoked by
+              the run() method. Defaults to None, meaning nothing is called.
+            args (tuple): is a list or tuple of arguments for the target
+              invocation. Defaults to ().
+        """
+        if (platform.system() == 'Linux'):
+            thread = threading.Thread(target=target, args=args)
+            thread.deamon = True
+            thread.start()
+            return thread
+        else:
+            thread = mp.Process(target=target, args=args)
+            thread.start()
+            return thread
 
     @staticmethod
     def _transcription_worker(conn,
@@ -948,37 +979,36 @@ class AudioToTextRecorder:
         """
         Processes audio data to detect wake words.
         """
-        match self.wakeword_backend:
-            case 'pvp' | 'pvporcupine':
-                pcm = struct.unpack_from(
-                    "h" * self.buffer_size,
-                    data
-                )
-                porcupine_index = self.porcupine.process(pcm)
+        if self.wakeword_backend in {'pvp', 'pvporcupine'}:
+            pcm = struct.unpack_from(
+                "h" * self.buffer_size,
+                data
+            )
+            porcupine_index = self.porcupine.process(pcm)
+            if self.debug_mode:
+                print (f"wake words porcupine_index: {porcupine_index}")
+            return self.porcupine.process(pcm)
+
+        elif self.wakeword_backend in {'oww', 'openwakeword', 'openwakewords'}:
+            pcm = np.frombuffer(data, dtype=np.int16)
+            prediction = self.owwModel.predict(pcm)
+            max_score = -1
+            max_index = -1
+            wake_words_in_prediction = len(self.owwModel.prediction_buffer.keys())
+            self.wake_words_sensitivities
+            if wake_words_in_prediction:
+                for idx, mdl in enumerate(self.owwModel.prediction_buffer.keys()):
+                    scores = list(self.owwModel.prediction_buffer[mdl])
+                    if scores[-1] >= self.wake_words_sensitivity and scores[-1] > max_score:
+                        max_score = scores[-1]
+                        max_index = idx
                 if self.debug_mode:
-                    print (f"wake words porcupine_index: {porcupine_index}")
-                return self.porcupine.process(pcm)
-
-            case 'oww' | 'openwakeword' | 'openwakewords':
-                pcm = np.frombuffer(data, dtype=np.int16)
-                prediction = self.owwModel.predict(pcm)
-                max_score = -1
-                max_index = -1
-                wake_words_in_prediction = len(self.owwModel.prediction_buffer.keys())
-                self.wake_words_sensitivities
-                if wake_words_in_prediction:
-                    for idx, mdl in enumerate(self.owwModel.prediction_buffer.keys()):
-                        scores = list(self.owwModel.prediction_buffer[mdl])
-                        if scores[-1] >= self.wake_words_sensitivity and scores[-1] > max_score:
-                            max_score = scores[-1]
-                            max_index = idx
-                    if self.debug_mode:
-                        print (f"wake words oww max_index, max_score: {max_index} {max_score}")
-                    return max_index  
-                else:
-                    if self.debug_mode:
-                        print (f"wake words oww_index: -1")
-                    return -1
+                    print (f"wake words oww max_index, max_score: {max_index} {max_score}")
+                return max_index  
+            else:
+                if self.debug_mode:
+                    print (f"wake words oww_index: -1")
+                return -1
 
         if self.debug_mode:        
             print("wake words no match")
@@ -1328,14 +1358,16 @@ class AudioToTextRecorder:
 
                     # Stop the recording if silence is detected after speech
                     if self.stop_recording_on_voice_deactivity:
+                        is_speech = (
+                            self._is_silero_speech(data) if self.silero_deactivity_detection
+                            else self._is_webrtc_speech(data, True)
+                        )
 
-                        if not self._is_webrtc_speech(data, True):
-
+                        if not is_speech:
                             # Voice deactivity was detected, so we start
                             # measuring silence time before stopping recording
                             if self.speech_end_silence_start == 0:
                                 self.speech_end_silence_start = time.time()
-
                         else:
                             self.speech_end_silence_start = 0
 

{RealtimeSTT-0.2.0/RealTimeSTT.egg-info → RealtimeSTT-0.2.2/RealtimeSTT.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: RealtimeSTT
-Version: 0.2.0
+Version: 0.2.2
 Summary: A fast Voice Activity Detection and Transcription System
 Home-page: https://github.com/KoljaB/RealTimeSTT
 Author: Kolja Beigel
@@ -12,16 +12,6 @@ Classifier: Operating System :: OS Independent
 Requires-Python: >=3.6
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: PyAudio==0.2.14
-Requires-Dist: faster-whisper==1.0.2
-Requires-Dist: pvporcupine==1.9.5
-Requires-Dist: webrtcvad==2.0.10
-Requires-Dist: halo==0.0.31
-Requires-Dist: torch==2.3.1
-Requires-Dist: torchaudio==2.3.1
-Requires-Dist: scipy==1.12.0
-Requires-Dist: websockets==v12.0
-Requires-Dist: openwakeword==0.6.0
 
 
 # RealtimeSTT
@@ -47,7 +37,7 @@ https://github.com/KoljaB/RealtimeSTT/assets/7604638/207cb9a2-4482-48e7-9d2b-072
 
 ### Updates
 
-Latest Version: v0.2.0
+Latest Version: v0.2.2
 
 See [release history](https://github.com/KoljaB/RealtimeSTT/releases).
 
@@ -89,24 +79,9 @@ Although it is possible to run RealtimeSTT with a CPU installation only (use a s
 ### GPU Support with CUDA (recommended)
 
 
-**Install RealtimeSTT with CUDA support**  
+### Updating PyTorch for CUDA Support
 
-*For Cuda 11.8:*  
-```bash
-pip install RealtimeSTT --index-url https://download.pytorch.org/whl/cu118
-```  
-
-
-*For Cuda 12.X:*  
-```bash
-pip install RealtimeSTT --index-url https://download.pytorch.org/whl/cu121
-```
-
-This command reinstalls PyTorch and Torchaudio with explicit support for CUDA.
-
-### Updating PyTorch Only for CUDA Support
-
-If you already have RealtimeSTT installed but want to upgrade your PyTorch installation to leverage GPU support with CUDA, follow these instructions based on your specific CUDA version. This process is useful if you initially installed RealtimeSTT with only CPU support and now wish to enhance its performance with CUDA capabilities.
+To upgrade your PyTorch installation to enable GPU support with CUDA, follow these instructions based on your specific CUDA version. This is useful if you wish to enhance the performance of RealtimeSTT with CUDA capabilities.
 
 #### For CUDA 11.8:
 To update PyTorch and Torchaudio to support CUDA 11.8, use the following commands:
@@ -122,7 +97,7 @@ To update PyTorch and Torchaudio to support CUDA 12.X, execute the following:
 pip install torch==2.3.1+cu121 torchaudio==2.3.1 --index-url https://download.pytorch.org/whl/cu121
 ```
 
-Replace `2.3.1` with the version of CUDA that matches your system and requirements.
+Replace `2.3.1` with the version of PyTorch that matches your system and requirements.
 
 ### Steps That Might Be Necessary Before
 
@@ -373,6 +348,8 @@ When you initialize the `AudioToTextRecorder` class, you have various options to
 
 - **silero_use_onnx** (bool, default=False): Enables usage of the pre-trained model from Silero in the ONNX (Open Neural Network Exchange) format instead of the PyTorch format. Default is False. Recommended for faster performance.
 
+- **silero_deactivity_detection** (bool, default=False): Enables the Silero model for end-of-speech detection. More robust against background noise. Utilizes additional GPU resources but improves accuracy in noisy environments. When False, uses the default WebRTC VAD, which is more sensitive but may continue recording longer due to background sounds.
+
 - **webrtc_sensitivity** (int, default=3): Sensitivity for the WebRTC Voice Activity Detection engine ranging from 0 (least aggressive / most sensitive) to 3 (most aggressive, least sensitive). Default is 3.
 
 - **post_speech_silence_duration** (float, default=0.2): Duration in seconds of silence that must follow speech before the recording is considered to be completed. This ensures that any brief pauses during speech don't prematurely end the recording.

{RealtimeSTT-0.2.0/RealTimeSTT.egg-info → RealtimeSTT-0.2.2/RealtimeSTT.egg-info}/SOURCES.txt

@@ -1,11 +1,6 @@
 LICENSE
 README.md
 setup.py
-RealTimeSTT.egg-info/PKG-INFO
-RealTimeSTT.egg-info/SOURCES.txt
-RealTimeSTT.egg-info/dependency_links.txt
-RealTimeSTT.egg-info/requires.txt
-RealTimeSTT.egg-info/top_level.txt
 RealtimeSTT/__init__.py
 RealtimeSTT/audio_recorder.py
 RealtimeSTT.egg-info/PKG-INFO

{RealtimeSTT-0.2.0/RealTimeSTT.egg-info → RealtimeSTT-0.2.2/RealtimeSTT.egg-info}/requires.txt

@@ -1,5 +1,5 @@
 PyAudio==0.2.14
-faster-whisper==1.0.2
+faster-whisper==1.0.3
 pvporcupine==1.9.5
 webrtcvad==2.0.10
 halo==0.0.31

{RealtimeSTT-0.2.0 → RealtimeSTT-0.2.2}/setup.py

@@ -9,7 +9,7 @@ with open('requirements.txt') as f:
 
 setuptools.setup(
     name="RealtimeSTT",
-    version="0.2.0",
+    version="0.2.2",
     author="Kolja Beigel",
     author_email="kolja.beigel@web.de",
     description="A fast Voice Activity Detection and Transcription System",