# pip install sounddevice
from supertonic import TTS
import sounddevice as sd
tts = TTS(auto_download=True)
style = tts.get_voice_style(voice_name='M1')
text = "This morning, I took a walk in the park, and the sound of the birds and the breeze was so pleasant that I stopped for a long time just to listen."
wav, duration = tts.synthesize(text, voice_style=style, lang='en')
print(f'Generated audio (duration: {duration[0]:.2f} seconds)')
# # Inspect the waveform shape
# print(f"Waveform shape: {wav.shape}")
# print(f"Waveform dtype: {wav.dtype}")
# Ensures the array is 1D, which is the standard format for mono audio playback in libraries like `sounddevice`. This prevents shape mismatches during audio output, where sounddevice expects a 1D array for single-channel audio.
# Without squeezing, sounddevice is unable to interpret the shape correctly, leading to errors or unexpected behavior during playback.
wav = wav.squeeze()
# Play the audio using sounddevice
sd.play(wav.astype("float32"))
sd.wait()
