app.py

import gradio as gr
from transformers import pipeline

# Load text-to-speech model
tts_title = "Text to Speech Translation"
tts_examples = ["I love learning machine learning", "How do you do?"]
tts_demo = gr.Interface.load(
    "huggingface/facebook/fastspeech2-en-ljspeech",
    title=tts_title,
    examples=tts_examples,
    description="Give me something to say!",
)

# Load emotion classification model
emotion_model_checkpoint = "MuntasirHossain/RoBERTa-base-finetuned-emotion"
emotion_model = pipeline("text-classification", model=emotion_model_checkpoint)

def classify_emotion_and_speech(text=""):
    # Emotion classification
    emotion_label = emotion_model(text)[0]["label"]

    # Adjust speech synthesis parameters based on emotion_label.
    # Customize this part based on the emotion_label.

    # Replace the following line with your desired text-to-speech model and parameters.
    speech_output = f"Emotion: {emotion_label}, Text: {text}"

    return {"emotion_label": emotion_label, "audio": speech_output}

emotion_title = "Texts Expressing Emotion with Speech"
emotion_description = "This AI model classifies texts expressing human emotion and converts them into speech."
emotion_examples = [["He is very happy today", "Free Palestine"]]

# Define a theme for the Gradio interface
theme = {
    "container": {
        "background-color": "#007bff",
        "color": "#fff",
        "padding": "20px",
    },
    "textbox": {
        "background-color": "#fff",
        "border-radius": "5px",
        "padding": "10px",
        "margin-bottom": "10px",
    },
    "button": {
        "background-color": "#007bff",
        "color": "#fff",
        "padding": "10px",
        "border-radius": "5px",
        "cursor": "pointer",
    },
    "label": {
        "color": "#fff",
    },
}

combined_demo = gr.Interface(
    fn=classify_emotion_and_speech,
    inputs="textbox",
    outputs=["text", "audio"],
    title=emotion_title,
    theme=theme,
    description=emotion_description,
    examples=emotion_examples,
)

# Combine both demos into a Tabbed Interface
combined_demo_tabbed = gr.TabbedInterface([tts_demo, combined_demo], ["Text to Speech", "Texts Expressing Emotion with Speech"])

if __name__ == "__main__":
    combined_demo_tabbed.launch()