stream_voice.py

Source: sunholo/senses/stream_voice.py

Functions

setup_tts_subparser(subparsers)

Sets up an argparse subparser for the 'tts' command.

Args: subparsers: The subparsers object from argparse.ArgumentParser().

tts_command(args)

Executes the TTS command based on parsed arguments.

Args: args: The parsed command-line arguments.

Classes

StreamingTTS

Example usage

def sample_text_stream(): sentences = [ "Hello, this is a test of streaming text to speech.", "Each sentence will be converted to audio separately.", "This allows for lower latency in long-form text to speech conversion." ] for sentence in sentences: yield sentence time.sleep(0.5) # Simulate delay between text chunks

Initialize and run

tts = StreamingTTS() tts.process_text_stream(sample_text_stream())

• del(self)

  • Cleanup method to ensure stream is closed.

• init(self)

  • Initialize self. See help(type(self)) for accurate signature.

• _apply_fade(self, audio: numpy.ndarray, fade_duration: float, fade_in: bool = True, fade_out: bool = True) -> numpy.ndarray

  • Apply fade in/out to audio with specified duration.

• _initialize_audio_device(self)

  • Initialize audio device with proper settings.

• _make_fade(self, length: int, fade_type: str = 'l') -> numpy.ndarray

  • Generate a fade curve of specified length and type.

• _play_audio_chunk(self, audio_chunk: numpy.ndarray, is_final_chunk: bool = False)

  • Play a single audio chunk with proper device handling.

• audio_player(self)

  • Continuously play audio chunks from the queue.

• generate_audio_stream(self, text)

  • Generate a stream of audio data from a text chunk. Returns audio in WAV format for streaming.

Args: text (str): Text to convert to speech

Yields: bytes: WAV-formatted audio data

• process_text_stream(self, text_generator)

  • Process incoming text stream and convert to audio.

• save_to_file(self, text_generator, output_path)

  • Save the audio to a WAV file with minimal fading.

• set_language(self, language_code: str)

  • Set the language for text-to-speech conversion.

Args: language_code: Language code in BCP-47 format (e.g., 'en-US', 'es-ES', 'fr-FR')

• set_voice(self, voice_name: str)
  • Set the language for text-to-speech conversion.

Args: language_code: Language code in BCP-47 format (e.g., 'en-US', 'es-ES', 'fr-FR')

• set_voice_gender(self, gender: str)
  • Set the voice gender for text-to-speech conversion.

Args: gender: One of 'NEUTRAL', 'MALE', or 'FEMALE'

• text_to_audio(self, text)
  • Convert text chunk to audio bytes using Google Cloud TTS.