import gradio as gr from gtts import gTTS import time import difflib import tempfile import os import speech_recognition as sr # Function to play the text (optional) def play_text(text): tts = gTTS(text=text, lang='hi', slow=False) temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.mp3') tts.save(temp_file.name) os.system(f"start {temp_file.name}") # Windows return "✅ Text is being read out. Please listen and read it yourself." # Function to transcribe user's audio and compare with the original text def transcribe_audio(audio, original_text): recognizer = sr.Recognizer() with sr.AudioFile(audio) as source: audio_data = recognizer.record(source) try: start_time = time.time() # Using Google Speech Recognition (supports Hindi) transcription = recognizer.recognize_google(audio_data, language="hi-IN") end_time = time.time() # Calculate Accuracy original_words = original_text.strip().split() transcribed_words = transcription.strip().split() matcher = difflib.SequenceMatcher(None, original_words, transcribed_words) accuracy = round(matcher.ratio() * 100, 2) # Calculate speed duration = end_time - start_time # time to process (not speaking time) speed = round(len(transcribed_words) / duration, 2) # words per second # Compare words and highlight mistakes wrong_words = [] for i, word in enumerate(original_words): if i >= len(transcribed_words) or word != transcribed_words[i]: wrong_words.append(f"🔴 {word}") result = { "📝 Transcribed Text": transcription, "🎯 Accuracy (%)": accuracy, "⏱️ Speaking Speed (words/sec)": speed, "❌ Incorrect Words": ' '.join(wrong_words) if wrong_words else "None" } return result except Exception as e: return {"error": str(e)} # Gradio App with gr.Blocks() as app: gr.Markdown("## 🗣️ Hindi Reading & Pronunciation Practice App") with gr.Row(): input_text = gr.Textbox(label="Paste Hindi Text Here", placeholder="यहाँ हिंदी टेक्स्ट लिखें...") play_button = gr.Button("🔊 Listen to Text") play_button.click(play_text, inputs=[input_text], outputs=[]) gr.Markdown("### 🎤 Now upload or record yourself reading the text aloud below:") audio_input = gr.Audio(type="filepath", label="Upload or Record Your Voice") submit_button = gr.Button("✅ Submit Recording for Checking") output = gr.JSON(label="Results") submit_button.click(transcribe_audio, inputs=[audio_input, input_text], outputs=[output]) # Launch the app app.launch()