import os
import uuid

import gradio as gr
from supertonic import TTS

OUTPUT_DIR = "outputs"
os.makedirs(OUTPUT_DIR, exist_ok=True)

# The first run downloads the Supertonic model assets automatically.
tts = TTS(auto_download=True)

SUPPORTED_LANGUAGES = {
    "Italiano": "it",
    "Inglese": "en",
    "Spagnolo": "es",
    "Francese": "fr",
    "Tedesco": "de",
    "Portoghese": "pt",
}

VOICES = ["M1", "M2", "M3", "M4", "M5", "F1", "F2", "F3", "F4", "F5"]


def generate_audio(text: str, language_label: str, voice: str):
    """Generate a local WAV file with Supertonic TTS."""
    if not text or not text.strip():
        return None, "Inserisci un testo prima di generare l'audio."

    lang = SUPPORTED_LANGUAGES.get(language_label, "it")
    style = tts.get_voice_style(voice)

    wav, _duration = tts.synthesize(
        text.strip(),
        voice_style=style,
        lang=lang,
    )

    output_path = os.path.join(OUTPUT_DIR, f"supertonic_{uuid.uuid4().hex[:8]}.wav")
    tts.save_audio(wav, output_path)

    return output_path, f"Audio generato correttamente: {output_path}"


with gr.Blocks(title="Supertonic Local Demo") as demo:
    gr.Markdown(
        """
# Supertonic Local Demo

Mini pannello locale per generare audio con Supertonic TTS.

Scrivi un testo, scegli lingua e voce, poi genera un file WAV direttamente sul tuo PC.
        """
    )

    with gr.Row():
        with gr.Column():
            text_input = gr.Textbox(
                label="Testo",
                lines=8,
                value="Ciao, questo è un test vocale in italiano generato localmente con Supertonic.",
            )
            language_input = gr.Dropdown(
                choices=list(SUPPORTED_LANGUAGES.keys()),
                value="Italiano",
                label="Lingua",
            )
            voice_input = gr.Dropdown(
                choices=VOICES,
                value="M1",
                label="Voce",
            )
            generate_button = gr.Button("Genera audio")

        with gr.Column():
            audio_output = gr.Audio(label="Audio generato", type="filepath")
            status_output = gr.Textbox(label="Stato")

    generate_button.click(
        fn=generate_audio,
        inputs=[text_input, language_input, voice_input],
        outputs=[audio_output, status_output],
    )


if __name__ == "__main__":
    demo.launch(server_name="0.0.0.0")