Create app.py

app.py

@@ -0,0 +1,86 @@
+import gradio as gr
+import speech_recognition as sr
+from pydub import AudioSegment
+import os
+from gtts import gTTS
+
+def transcribe_audio(audio):
+    recognizer = sr.Recognizer()
+    audio = AudioSegment.from_file(audio)
+    audio.export("temp.wav", format="wav")
+    with sr.AudioFile("temp.wav") as source:
+        audio_data = recognizer.record(source)
+        try:
+            text = recognizer.recognize_google(audio_data)
+        except sr.UnknownValueError:
+            text = "Sorry, I could not understand the audio."
+        except sr.RequestError:
+            text = "Sorry, there was an error with the speech recognition service."
+    os.remove("temp.wav")
+    return text
+
+def speak_text(text):
+    tts = gTTS(text)
+    tts.save("question.mp3")
+    return "question.mp3"
+
+questions = [
+    {"context": "Let's talk about the story, My Dad's birthday party. Today is your dad’s birthday. So you and your family will have a party tonight.", "question": "What will your mom do for the party?", "label": "Transcription of Mom's action", "answer": "She will cook the dinner."},
+    {"context": "Look at the page 12 and 13.", "question": "What is she doing?", "label": "Transcription of Mom's action", "answer": "She is baking a cake."},
+    {"context": "How about your brother?", "question": "What will he do for the party?", "label": "Transcription of Brother's action", "answer": "He will decorate the room."},
+    {"context": "Look at the picture.", "question": "What is he doing?", "label": "Transcription of Brother's action", "answer": "He is hanging balloons."},
+    {"context": "Okay, next,", "question": "How about you? What will you do for the party?", "label": "Transcription of Your action", "answer": "I will set the table."},
+    {"context": "Let's move on to the story 'Owls are special'. Owls are nocturnal.", "question": "When do they sleep?", "label": "Transcription of sleep time", "answer": "They sleep during the day."},
+    {"context": "Look at the page 21, they have very special eyes.", "question": "So, what can they do with their special eyes?", "label": "Transcription of eye ability", "answer": "They can see in the dark."},
+    {"context": "Now, these questions are based on the story 'I will go shopping'. You have many things to buy. First, you need to buy a book. So, you will go to the bookstore. The bookstore is far away.", "question": "How will you go to the bookstore?", "label": "Transcription of transport method", "answer": "I will take the bus."},
+    {"context": "Now you need to buy some bread.", "question": "Where will you go?", "label": "Transcription of place", "answer": "I will go to the bakery."},
+    {"context": "Let's talk about the story 'Guide dogs'. Blind people can't see. But, they can still do things.", "question": "How can they do?", "label": "Transcription of how blind people do things", "answer": "They use their other senses."},
+    {"context": "Guide dogs help blind people with everything. For example, they bring things for them. They help them. They open doors for them.", "question": "What else can they do for them?", "label": "Transcription of other abilities", "answer": "They guide them while walking."}
+]
+
+current_question = 0
+responses = []
+
+def next_question():
+    global current_question
+    if current_question < len(questions):
+        context = questions[current_question].get("context", "")
+        question = questions[current_question]["question"]
+        full_text = f"{context} {question}"
+        question_audio = speak_text(full_text)
+        current_question += 1
+        return gr.update(value=question_audio, visible=True), gr.update(visible=True), questions[current_question-1]["label"], gr.update(visible=True), gr.update(visible=False)
+    else:
+        final_results = evaluate_responses()
+        return gr.update(visible=False), gr.update(visible=False), gr.update(visible=False), gr.update(visible=False), gr.update(value=final_results, visible=True)
+
+def save_response(audio):
+    transcription = transcribe_audio(audio)
+    responses.append(transcription)
+    return transcription
+
+def evaluate_responses():
+    result = "<h2>Your Responses:</h2><br>"
+    for i, question in enumerate(questions):
+        user_response = responses[i] if i < len(responses) else "No response"
+        result += f"<b>Q:</b> {question['question']}<br><b>Your Answer:</b> {user_response}<br><br>"
+    return result
+
+with gr.Blocks() as demo:
+    gr.Markdown("### Interactive Questions")
+
+    question_audio = gr.Audio(label="Question", visible=False)
+    audio_input = gr.Audio(type="filepath", label="Your answer", visible=True)
+    transcription_output = gr.Textbox(label="Transcription", visible=True)
+    btn_next = gr.Button("Next", visible=True)
+    final_output = gr.HTML(visible=False)
+
+    def load_first_question():
+        return next_question()
+
+    demo.load(load_first_question, outputs=[question_audio, audio_input, transcription_output, btn_next, final_output])
+
+    btn_next.click(next_question, outputs=[question_audio, audio_input, transcription_output, btn_next, final_output])
+    audio_input.change(save_response, inputs=audio_input, outputs=transcription_output)
+
+    demo.launch()