Update app.py

app.py

@@ -1,78 +1,61 @@
 # import part
 import streamlit as st
 from transformers import pipeline
-from gtts import gTTS
-import os
 
 # function part
 # img2text
 def img2text(url):
-    try:
-        image_to_text_model = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")
-        text = image_to_text_model(url)[0]["generated_text"]
-        # Make the caption more fun and creative
-        fun_caption = f"✨ Wow! Look at this! It’s a picture of {text.lower()}. Let’s turn it into a fun story! ✨"
-        return fun_caption
-    except Exception as e:
-        st.error(f"Oops! Something went wrong while looking at your picture. Please try again! 🪄")
-        return None
+    image_to_text_model = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")
+    text = image_to_text_model(url)[0]["generated_text"]
+    # Make the caption more fun and descriptive
+    fun_caption = f"Wow! {text.capitalize()}! 🌟"
+    return fun_caption
 
 # text2story
 def text2story(text):
-    try:
-        story_generator = pipeline("text-generation", model="gpt2")
-        # Add a playful prompt to guide the story generation
-        prompt = f"Once upon a time, there was {text}. "
-        story = story_generator(prompt, max_length=95, num_return_sequences=1)[0]["generated_text"]
-        # Make the story more fun by adding a happy ending
-        fun_story = story + " And they all lived happily ever after! 🌈✨"
-        return fun_story
-    except Exception as e:
-        st.error(f"Oops! Something went wrong while creating your story. Please try again! 🪄")
-        return None
+    # Use a text-generation model to create a fun and realistic story
+    story_generator = pipeline("text-generation", model="gpt2")
+    # Add a prompt to guide the story generation (avoid magical elements)
+    prompt = f"Write a fun and realistic story for kids based on this: {text}. Keep it simple and under 95 words."
+    story = story_generator(prompt, max_length=95, num_return_sequences=1)[0]["generated_text"]
+    # Ensure the story is simple and fun for kids
+    simple_story = story[:95]  # Limit to 95 words
+    return simple_story
 
 # text2audio
 def text2audio(story_text):
-    try:
-        tts = gTTS(text=story_text, lang='en')
-        audio_file = "story_audio.mp3"
-        tts.save(audio_file)
-        return audio_file
-    except Exception as e:
-        st.error(f"Oops! Something went wrong while turning your story into audio. Please try again! 🪄")
-        return None
+    # Use a TTS model to convert text to audio
+    tts = pipeline("text-to-speech", model="facebook/fastspeech2-en-ljspeech")
+    audio_data = tts(story_text)
+    return audio_data
 
 # main part
-st.set_page_config(page_title="Story Maker", page_icon="🦄")
-st.header("🌟 Story Maker: Turn Your Picture into a Fun Story! 🌟")
-uploaded_file = st.file_uploader("📷 Choose a picture to create a fun story...", type=["jpg", "jpeg", "png"])
+st.set_page_config(page_title="Story Explorer", page_icon="🦜")
+st.header("Story Explorer: Turn Your Picture into a Fun Story! 🎨📖")
+
+uploaded_file = st.file_uploader("Choose a picture...", type=["jpg", "png", "jpeg"])
 
 if uploaded_file is not None:
     bytes_data = uploaded_file.getvalue()
     with open(uploaded_file.name, "wb") as file:
         file.write(bytes_data)
 
-    st.image(uploaded_file, caption="Your picture!", use_container_width=True)
+    st.image(uploaded_file, caption="Your Picture", use_container_width=True)
 
     # Stage 1: Image to Text
-    st.write("✨ Let’s see what’s in your picture... ✨")
+    st.text('Let’s explore your picture! 🧐✨')
     scenario = img2text(uploaded_file.name)
-    if scenario:
-        st.write(scenario)
+    st.write(f"**Here’s what I see in your picture:** {scenario}")
+
+    # Stage 2: Text to Story
+    st.text('Creating a fun story for you! 📖')
+    story = text2story(scenario)
+    st.write(f"**Here’s your story:** {story}")
 
-        # Stage 2: Text to Story
-        st.write("🧙‍♂️ Turning your picture into a story... 🧙‍♀️")
-        story = text2story(scenario)
-        if story:
-            st.write("📖 Here’s your story:")
-            st.write(story)
+    # Stage 3: Story to Audio data
+    st.text('Turning your story into audio... 🎧')
+    audio_data = text2audio(story)
 
-            # Stage 3: Story to Audio
-            st.write("🎤 Getting ready to tell your story... 🎤")
-            audio_file = text2audio(story)
-            if audio_file:
-                # Play button
-                if st.button("🎧 Listen to Your Story!"):
-                    st.audio(audio_file, format="audio/mp3")
-                # Clean up the audio file after playing
-                os.remove(audio_file)
+    # Play button
+    if st.button("Play Audio"):
+        st.audio(audio_data['audio'], format="audio/wav", start_time=0, sample_rate=audio_data['sampling_rate'])