text_to_speech.py

import os
import time
from io import BytesIO
import pygame
from elevenlabs import ElevenLabs
from elevenlabs import VoiceSettings

def text_to_speech_elevenlabs(text):
    """
    Convert text to speech using ElevenLabs API and play the audio.
    """
    pygame.mixer.init()

    try:
        elevenclient = ElevenLabs(api_key=os.getenv("ELEVEN_API_KEY"))
        response = elevenclient.text_to_speech.convert(
            voice_id="MF3mGyEYCl7XYWbV9V6O",
            output_format="mp3_22050_32",
            text=text,
            model_id="eleven_turbo_v2",
            voice_settings=VoiceSettings(stability=0.0, similarity_boost=1.0, style=0.0, use_speaker_boost=True),
        )

        # Write the audio data to a BytesIO object
        audio_data = BytesIO()
        for chunk in response:
            if chunk:
                audio_data.write(chunk)

        if audio_data.getbuffer().nbytes == 0:
            raise ValueError("Audio data is empty.")

        # Play the audio using pygame
        audio_data.seek(0)
        pygame.mixer.music.load(audio_data)
        pygame.mixer.music.play()

        # Wait for the audio to finish playing
        while pygame.mixer.music.get_busy():
            time.sleep(0.1)

        return "Audio played successfully"

    except Exception as e:
        print(f"An error occurred: {str(e)}")
        raise