app.py

import gradio as gr
import json
#from bark import SAMPLE_RATE, generate_audio, preload_models
#from bark.generation import SUPPORTED_LANGS
#from share_btn imposrt community_icon_html, loading_icon_html, share_js
#from flask import Flask, jsonify
 
DEBUG_MODE = False
TITLE = "《★~ N.N.F. Text To Speech V2 ~★ 》"
#FAVICON = "https://aeiljuispo.cloudimg.io/v7/https://cdn-uploads.huggingface.co/production/uploads/65b453089151075ad87382ea/9NpWGb4xBrZovX1o7DF-5.jpeg"

INPUT_TEXT_PROMPT_EXAMPLE="""
Não sou nada,
não serei nada,
não posso querer ser nada.
À parte isso, tenho em mim todos os sonhos do mundo.
"""

DECRIPTION = """
This is a simple Text To Speech (TTS) model that uses the XTTS model from Hugging Face.
You can use this model to generate audio from text in multiple languages.
It can be used as an API or as a standalone application.
"""

def predict(text, lang, audio, request: gr.Request):
    output_text = {"verdict ": "SUCCESS"}  # Initialize as a dictionary
    output_text["Text"] = text
    output_text["Language"] = lang
    
    if request:
        # Convert headers to a dictionary and include them in the output_text
        output_text["headers"] = dict(request.headers.items())

    output_text_json = json.dumps(output_text)
    return output_text_json

io = gr.Interface(
    fn=predict,
    inputs=[
        #definition of the input text to parse
        gr.Textbox(
            label="Text Prompt",
            info="One or two sentences at a time is better",
            value=INPUT_TEXT_PROMPT_EXAMPLE,
        ),
        gr.Dropdown(
            label="Language",
            info="Select an output language for the reader",
            choices=[
                "en",
                "es",
                "fr",
                "it",
                "pt",
                "nl",
            ],
            #max_choices=1,
            value="pt",
        ),
        gr.Audio(
            #label="Reference Audio",
            #info="Click on the ✎ button to upload your own target speaker audio",
            #type="filepath",
            #value="examples/female.wav",
        ),
    ],
    outputs=[
        #gr.Video(label="Waveform Visual"),
        "json",
    ],
    title=TITLE,
    description=DECRIPTION,
    #favicon_path=FAVICON,
    #article=article,
    #cache_examples=False,
    #examples=examples,
).queue().launch(share=True)