import gradio as gr from gtts import gTTS import os from datetime import datetime import tempfile from transformers import pipeline # Emotion detection model classifier = pipeline("text-classification", model="j-hartmann/emotion-english-distilroberta-base") # Chakra data (simplified example; extend as needed) chakra_data = { "Root": { "Emotion": ["fear", "anxiety"], "Purpose": "Grounding, survival", "Why_Imbalanced": "Feeling unsafe or insecure", "How_to_Balance": "Practice grounding, connect with nature", "Affirmation": "I am safe and secure." }, "Heart": { "Emotion": ["sadness", "grief"], "Purpose": "Love, compassion", "Why_Imbalanced": "Holding on to past pain or resentment", "How_to_Balance": "Practice forgiveness and self-love", "Affirmation": "I am open to love and kindness." }, "Throat": { "Emotion": ["frustration"], "Purpose": "Communication and expression", "Why_Imbalanced": "Fear of speaking out or not being heard", "How_to_Balance": "Speak your truth and express yourself", "Affirmation": "I speak my truth with clarity and courage." } } # Detect emotion def detect_emotion_nlp(text): result = classifier(text) return result[0]['label'].lower() # Analyze input def analyze_from_text_nlp(journal_text): detected_emotion = detect_emotion_nlp(journal_text) for chakra, data in chakra_data.items(): if detected_emotion in [e.lower() for e in data["Emotion"]]: return ( f"### Chakra Affected: {chakra}", f"**Purpose**: {data['Purpose']}", f"**Why Imbalanced**: {data['Why_Imbalanced']}",**How to Balance**: {data['How_to_Balance']}", data["Affirmation"] ) return ("No chakra affected", "", "", "") # Play affirmation def play_affirmation_audio(affirmation_text): tts = gTTS(text=affirmation_text, lang='en') tmp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") tts.save(tmp_file.name) return tmp_file.name # Gradio UI with gr.Blocks() as demo: gr.Markdown("## Shunya - Chakra Emotion Analyzer") with gr.Tab("Analyze"): journal_input = gr.Textbox(label="Enter your journal or emotion") analyze_btn = gr.Button("Analyze Emotion") chakra_output = gr.Markdown() purpose_output = gr.Markdown() balance_output = gr.Markdown() affirmation_output = gr.Textbox(label="Affirmation", interactive=False) analyze_btn.click( analyze_from_text_nlp, inputs=journal_input, outputs=[chakra_output, purpose_output, balance_output, affirmation_output] ) with gr.Tab("Affirmation"): gr.Markdown("## Listen to Your Affirmation") affirmation_text = gr.Textbox(label="Affirmation", interactive=False) play_audio_btn = gr.Button("Play Affirmation") affirmation_audio = gr.Audio(label="Audio", interactive=False, type="filepath") play_audio_btn.click(play_affirmation_audio, inputs=affirmation_text, outputs=affirmation_audio) demo.launch()