webscout 6.3__py3-none-any.whl → 6.5__py3-none-any.whl

webscout/Provider/TTS/murfai.py

@@ -0,0 +1,139 @@
+import time
+import requests
+import pathlib
+from io import BytesIO
+from urllib.parse import urlencode
+from playsound import playsound
+from webscout import exceptions
+from webscout.AIbase import TTSProvider
+from webscout.Litlogger import LitLogger, LogFormat, ColorScheme
+from webscout.litagent import LitAgent
+from concurrent.futures import ThreadPoolExecutor, as_completed
+from . import utils
+
+class MurfAITTS(TTSProvider):
+    """Text-to-speech provider using the MurfAITTS API."""
+    # Request headers
+    headers: dict[str, str] = {
+        "User-Agent": LitAgent().random()
+    }
+    cache_dir = pathlib.Path("./audio_cache")
+    all_voices: dict[str, str] = {"Hazel": "en-UK-hazel"}
+
+    def __init__(self, timeout: int = 20, proxies: dict = None):
+        """Initializes the MurfAITTS TTS client."""
+        self.session = requests.Session()
+        self.session.headers.update(self.headers)
+        if proxies:
+            self.session.proxies.update(proxies)
+        self.timeout = timeout
+        self.logger = LitLogger(
+            name="MurfAITTS",
+            format=LogFormat.MODERN_EMOJI,
+            color_scheme=ColorScheme.AURORA
+        )
+
+    def tts(self, text: str, voice: str = "Hazel", verbose:bool = True) -> str:
+        """Converts text to speech using the MurfAITTS API and saves it to a file."""
+        assert (
+            voice in self.all_voices
+        ), f"Voice '{voice}' not one of [{', '.join(self.all_voices.keys())}]"
+
+        filename = self.cache_dir / f"{int(time.time())}.mp3"
+
+        voice_id = self.all_voices[voice]
+
+        # Split text into sentences
+        sentences = utils.split_sentences(text)
+
+        # Function to request audio for each chunk
+        def generate_audio_for_chunk(part_text: str, part_number: int):
+            while True:
+                try:
+                    params: dict[str, str] = {
+                    "name": voice_id, 
+                    "text": part_text
+                    }
+                    encode_param: str = urlencode(params)
+                    response = self.session.get(f"https://murf.ai/Prod/anonymous-tts/audio?{encode_param}", headers=self.headers, timeout=self.timeout)
+                    response.raise_for_status()
+
+                    # Create the audio_cache directory if it doesn't exist
+                    self.cache_dir.mkdir(parents=True, exist_ok=True) 
+
+                    # Check if the request was successful
+                    if response.ok and response.status_code == 200:
+                        if verbose:
+                            self.logger.success(f"Chunk {part_number} processed successfully 🎉")
+                        return part_number, response.content
+                    else:
+                        if verbose:
+                            self.logger.warning(f"No data received for chunk {part_number}. Retrying...")
+                except requests.RequestException as e:
+                    if verbose:
+                        self.logger.error(f"Error for chunk {part_number}: {e}. Retrying... 🔄")
+                    time.sleep(1)
+        try:
+            # Using ThreadPoolExecutor to handle requests concurrently
+            with ThreadPoolExecutor() as executor:
+                futures = {executor.submit(generate_audio_for_chunk, sentence.strip(), chunk_num): chunk_num 
+                        for chunk_num, sentence in enumerate(sentences, start=1)}
+                
+                # Dictionary to store results with order preserved
+                audio_chunks = {}
+
+                for future in as_completed(futures):
+                    chunk_num = futures[future]
+                    try:
+                        part_number, audio_data = future.result()
+                        audio_chunks[part_number] = audio_data  # Store the audio data in correct sequence
+                    except Exception as e:
+                        if verbose:
+                            self.logger.error(f"Failed to generate audio for chunk {chunk_num}: {e} 🚨")
+
+            # Combine audio chunks in the correct sequence
+            combined_audio = BytesIO()
+            for part_number in sorted(audio_chunks.keys()):
+                combined_audio.write(audio_chunks[part_number])
+                if verbose:
+                    self.logger.debug(f"Added chunk {part_number} to the combined file.")
+
+            # Save the combined audio data to a single file
+            with open(filename, 'wb') as f:
+                f.write(combined_audio.getvalue())
+            if verbose:
+                self.logger.info(f"Final Audio Saved as {filename} 🔊")
+            return filename.as_posix()
+
+        except requests.exceptions.RequestException as e:
+            self.logger.critical(f"Failed to perform the operation: {e} 🚨")
+            raise exceptions.FailedToGenerateResponseError(
+                f"Failed to perform the operation: {e}"
+            )
+        
+    def play_audio(self, filename: str):
+        """
+        Plays an audio file using playsound.
+
+        Args:
+            filename (str): The path to the audio file.
+
+        Raises:
+            RuntimeError: If there is an error playing the audio.
+        """
+        try:
+            playsound(filename)
+        except Exception as e:
+            self.logger.error(f"Error playing audio: {e} 🔇")
+            raise RuntimeError(f"Error playing audio: {e}")
+
+# Example usage
+if __name__ == "__main__":
+    murfai = MurfAITTS()
+    text = "This is a test of the MurfAITTS text-to-speech API. It supports multiple sentences and advanced logging."
+
+    murfai.logger.info("Generating audio...")
+    audio_file = murfai.tts(text, voice="Hazel") 
+
+    murfai.logger.info("Playing audio...")
+    murfai.play_audio(audio_file)

webscout/Provider/TTS/parler.py

@@ -1,108 +1,135 @@
-import time
-from pathlib import Path
-from typing import Generator
-from playsound import playsound
-from webscout import exceptions
-from webscout.AIbase import TTSProvider
-
-from gradio_client import Client
-import os
-
-
-class ParlerTTS(TTSProvider):
-    """
-    A class to interact with the Parler TTS API through Gradio Client.
-    """
-
-    def __init__(self, timeout: int = 20, proxies: dict = None):
-        """Initializes the Parler TTS client."""
-        self.api_endpoint = "/gen_tts"
-        self.client = Client("parler-tts/parler_tts")  # Initialize the Gradio client
-        self.timeout = timeout
-        self.audio_cache_dir = Path("./audio_cache")
-
-    def tts(self, text: str, description: str = "", use_large: bool = False) -> str:
-        """
-        Converts text to speech using the Parler TTS API.
-
-        Args:
-            text (str): The text to be converted to speech.
-            description (str, optional): Description of the desired voice characteristics. Defaults to "".
-            use_large (bool, optional): Whether to use the large model variant. Defaults to False.
-
-        Returns:
-            str: The filename of the saved audio file.
-
-        Raises:
-            exceptions.FailedToGenerateResponseError: If there is an error generating or saving the audio.
-        """
-        filename = self.audio_cache_dir / f"{int(time.time())}.wav"
-
-        try:
-            result = self.client.predict(
-                text=text,
-                description=description,
-                use_large=use_large,
-                api_name=self.api_endpoint,
-            )
-
-            if isinstance(result, bytes):
-                audio_bytes = result
-            elif isinstance(result, str) and os.path.isfile(result):
-                with open(result, "rb") as f:
-                    audio_bytes = f.read()
-            else:
-                raise ValueError(f"Unexpected response from API: {result}")
-
-            self._save_audio(audio_bytes, filename)
-            return filename.as_posix()
-
-        except Exception as e:
-            raise exceptions.FailedToGenerateResponseError(
-                f"Error generating audio after multiple retries: {e}"
-            ) from e
-
-    def _save_audio(self, audio_data: bytes, filename: Path):
-        """Saves the audio data to a WAV file in the audio cache directory."""
-        try:
-            self.audio_cache_dir.mkdir(parents=True, exist_ok=True)
-            with open(filename, "wb") as f:
-                f.write(audio_data)
-
-        except Exception as e:
-            raise exceptions.FailedToGenerateResponseError(f"Error saving audio: {e}")
-
-    def play_audio(self, filename: str):
-        """
-        Plays an audio file using playsound.
-
-        Args:
-            filename (str): The path to the audio file.
-
-        Raises:
-            RuntimeError: If there is an error playing the audio.
-        """
-        try:
-            playsound(filename)
-        except Exception as e:
-            raise RuntimeError(f"Error playing audio: {e}")
-
-
-# Example usage
-if __name__ == "__main__":
-    parlertts = ParlerTTS()
-    text = (
-        "All of the data, pre-processing, training code, and weights are released "
-        "publicly under a permissive license, enabling the community to build on our work "
-        "and develop their own powerful models."
-    )
-    voice_description = (
-        "Laura's voice is monotone yet slightly fast in delivery, with a very close "
-        "recording that almost has no background noise."
-    )
-
-    print("Generating audio...")
-    audio_file = parlertts.tts(text, description=voice_description, use_large=False)
-
-    print("Playing audio...")
+import time
+from pathlib import Path
+from typing import Generator
+from playsound import playsound
+from webscout import exceptions
+from webscout.AIbase import TTSProvider
+from webscout.Litlogger import LitLogger, LogFormat, ColorScheme
+from gradio_client import Client
+import os
+
+
+class ParlerTTS(TTSProvider):
+    """
+    A class to interact with the Parler TTS API through Gradio Client.
+    """
+
+    def __init__(self, timeout: int = 20, proxies: dict = None):
+        """Initializes the Parler TTS client."""
+        self.api_endpoint = "/gen_tts"
+        self.client = Client("parler-tts/parler_tts")  # Initialize the Gradio client
+        self.timeout = timeout
+        self.audio_cache_dir = Path("./audio_cache")
+        self.logger = LitLogger(
+            name="ParlerTTS",
+            format=LogFormat.MODERN_EMOJI,
+            color_scheme=ColorScheme.AURORA
+        )
+
+    def tts(self, text: str, description: str = "", use_large: bool = False, verbose: bool = True) -> str:
+        """
+        Converts text to speech using the Parler TTS API.
+
+        Args:
+            text (str): The text to be converted to speech.
+            description (str, optional): Description of the desired voice characteristics. Defaults to "".
+            use_large (bool, optional): Whether to use the large model variant. Defaults to False.
+            verbose (bool, optional): Whether to log detailed information. Defaults to True.
+
+        Returns:
+            str: The filename of the saved audio file.
+
+        Raises:
+            exceptions.FailedToGenerateResponseError: If there is an error generating or saving the audio.
+        """
+        filename = self.audio_cache_dir / f"{int(time.time())}.wav"
+
+        try:
+            if verbose:
+                self.logger.info(f"Generating TTS with description: {description} 🎙️")
+            
+            result = self.client.predict(
+                text=text,
+                description=description,
+                use_large=use_large,
+                api_name=self.api_endpoint,
+            )
+
+            if isinstance(result, bytes):
+                audio_bytes = result
+            elif isinstance(result, str) and os.path.isfile(result):
+                with open(result, "rb") as f:
+                    audio_bytes = f.read()
+            else:
+                raise ValueError(f"Unexpected response from API: {result}")
+
+            self._save_audio(audio_bytes, filename)
+            
+            if verbose:
+                self.logger.success(f"Audio generated successfully: {filename} 🔊")
+            
+            return filename.as_posix()
+
+        except Exception as e:
+            self.logger.critical(f"Error generating audio: {e} 🚨")
+            raise exceptions.FailedToGenerateResponseError(
+                f"Error generating audio after multiple retries: {e}"
+            ) from e
+
+    def _save_audio(self, audio_data: bytes, filename: Path):
+        """
+        Saves the audio data to a WAV file in the audio cache directory.
+
+        Args:
+            audio_data (bytes): Audio data to save
+            filename (Path): Path to save the audio file
+
+        Raises:
+            exceptions.FailedToGenerateResponseError: If there is an error saving the audio.
+        """
+        try:
+            self.audio_cache_dir.mkdir(parents=True, exist_ok=True)
+            with open(filename, "wb") as f:
+                f.write(audio_data)
+            self.logger.debug(f"Audio saved to {filename} 💾")
+
+        except Exception as e:
+            self.logger.error(f"Error saving audio: {e} 🔇")
+            raise exceptions.FailedToGenerateResponseError(f"Error saving audio: {e}")
+
+    def play_audio(self, filename: str):
+        """
+        Plays an audio file using playsound.
+
+        Args:
+            filename (str): The path to the audio file.
+
+        Raises:
+            RuntimeError: If there is an error playing the audio.
+        """
+        try:
+            self.logger.info(f"Playing audio: {filename} 🎵")
+            playsound(filename)
+        except Exception as e:
+            self.logger.error(f"Error playing audio: {e} 🔇")
+            raise RuntimeError(f"Error playing audio: {e}")
+
+
+# Example usage
+if __name__ == "__main__":
+    parlertts = ParlerTTS()
+    text = (
+        "All of the data, pre-processing, training code, and weights are released "
+        "publicly under a permissive license, enabling the community to build on our work "
+        "and develop their own powerful models."
+    )
+    voice_description = (
+        "Laura's voice is monotone yet slightly fast in delivery, with a very close "
+        "recording that almost has no background noise."
+    )
+
+    parlertts.logger.info("Generating audio...")
+    audio_file = parlertts.tts(text, description=voice_description, use_large=False)
+
+    parlertts.logger.info("Playing audio...")
     parlertts.play_audio(audio_file)